IEEE Engineering in Medicine and Biology - vol 25 - nb 03 - May-June 2006

IEEE ENGINEERING IN

MEDICINE AND BIOLOGY Magazine VOLUME 25 • NUMBER 3 ■ http://EMB-Magazine.bme.uconn.edu ■ MAY/JUNE 2006

IEEE ENGINEERING IN VOLUME 25 • NUMBER 3 MAY/JUNE 2006 http://EMB-Magazine.bme.uconn.edu

MEDICINE AND BIOLOGY Magazine

Engineering Humanity

Themes 16

The Humanitarian Efforts of Biomedical Engineers Robert Malkin

18

Science, Engineering, and Humanity Richard R. Ernst

20

The Science of Volunteering Peter Creane

23

International Aid's Medical Equipment Training Program Billy Teninty

BACKGROUND IMAGE© DIGITAL VISION, LTD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

25

Senior Design for Persons with Disabilities John D. Enderle

30

Prescription for Success Greg Russell

32

Engineers Without Borders and Their Role in Humanitarian Relief Claes I Helgesson

36

A Low-Cost Solution to Rural Water Disinfection Charles Taflin

Features 38

The Evolution of Pacemakers Sandro A.P. Haddad, Richard P.M. Houben, and Wouter A. Serdijn

49

Locomotion Techniques for Robotic Colonoscopy Irwan Kassim, Louis Phee, Wan S. Ng, Feng Gong, Paolo Dario, and Charles A. Mosse

57

Noncontact Measurement of Breathing Function Ramya Murthy and Ioannis Pavlidis

68

Study of Facial Skin and Aural Temperature Eddie Y.K. Ng, Wiryani Muljo, and B. Stephen Wong

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

1

Departments 4 From the Editor The Bright Future of BME

6 Society News New Chapters, New Members

7 Student's Corner EMBS Students Surf More Than the Web

9 Book Reviews 12 BME Company Profiles CleveMed,Orbital Research, and Their Spinoffs

75 COMAR Report of COMAR Activities 2005–2006

76 Patents The U.S. Patent That Reached Around the World

77 Emerging Technologies Healthcare Applications of RF Identification

84 Engineering in Genomics Melatonin Administration Does Not Affect Isoproterenol-Induced LVH

88 Retrospectroscope Did Wheatstone Build a Bridge?

91 Cellular/Tissue Engineering Tissue Engineering and Eucomed News in Brief

92 GOLD EMBS Goes for the GOLD!

93 Conference Calendar

Mission Statement The Engineering in Medicine and Biology Society of the IEEE advances the application of engineering sciences and technology to medicine and biology, promotes the profession, and provides global leadership for the benefit of its members and humanity by disseminating knowledge, setting standards, fostering professional development, and recognizing excellence.

2

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

NOTES FOR CONTRIBUTORS IEEE Engineering in Medicine and Biology Magazine is a theme-article publication that covers the full range of fields within biomedical engineering (BME), with each issue covering one theme. Articles are written for technically knowledgeable readers who are not necessarily specialists in the theme topic. A sample list of theme topics of interest includes: biochemical engineering, biocontrols, bioinformatics, biomems, biomaterials, biomechanics, biosignal processing, biotechnology, cellular and tissue engineering, clinical engineering, imaging and image processing, information technology, instrumentation, sensors and measurements, micro- and nanotechnolgy, neural systems and engineering, physiological systems modeling, proteomics, radiology, rehabilitation engineerNOTES FORinCONTRIBUTORS ing, robotics surgery, and telemedicine. In addition to the theme articles, which are invited contributions, the magazine also publishes unsolicited features Coming Attractions that areissues of interest a broad segment IEEE Engineering in Medicine and Future of thetomagazine will haveofthe themes of Teaching Engineering Biology Magazine readers. Tissue Engineering, Biotechnology, and Wearable in Medicine and Biology. IEEE Engineering Medicine andonBiology also publishes 20 Sensors/Systems and in Their Impact BME. Magazine Contributions on themeover topics regularly scheduled columnsarticles for readers interested in industry, academia, and are invited. Other technical and feature stories of interest to biomedgovernment. are peer reviewed andare written by experts in the field. ical engineersAll arearticles also welcome. All articles submitted anonymously for On the magazine comprehensive, in-depth tutorial, peeroccasion, review. Letters to the publishes editor, notes, commentaries, and review, other pieces of and survey articles. Letters to the editor, notes, other related pieces personal opinion will be published as such. Wecommentaries, also seek pressand releases of personal in opinion will be published as such. We also seek press releases to activities your company, organization, or school. related to activities your company, school. Manuscripts areinONLY accepted organization, in electronic or format through Manuscript Manuscripts aresite ONLY accepted in electronic format through Manuscript Central at the Web http://embs-ieee.manuscriptcentral.com. Instructions for Central Web and site how http://embs-ieee.manuscriptcentral.com. Instructions for creatingatanthe account to electronically submit a manuscript are available creating an account to original electronically submit aormanuscript available at at the Web site. Doand nothow send submissions revisions are directly to the the Web site. Do Ifnot send or revisions directly to the editorEditor-in-Chief. you areoriginal unable submissions to submit your contribution electronically or in-chief. If you are to submit yourplease contribution or have queshave questions onunable manuscripts style, contactelectronically the Editor-in-Chief: Dr. tions on Enderle, manuscripts style, please contact the editor-in-chief: D. Enderle, John D. Biomedical Engineering Director, UniversityJohn of Connecticut, Program DirectorRoad, for Biomedical Connecticut, 260 Glenbrook Storrs, CTEngineering, 06269-2247.University Voice: +1of860 486 5521.Room Fax: 217, 260486 Glenbrook Road, Storrs, CT 06269-2247 USA. Voice: +1 860 486 5521. +1 860 2500. E-mail: [email protected]. Fax:As +1 860 2500. E-mail: [email protected]. per 486 IEEE policy on standards for publications, review and editorial As per policy are on part standards publications, editorial handling of IEEE manuscripts of the for paper submision review process and to guarantee handling of manuscripts are part of the paper submision process to guarantee quality control. quality control. Make the Deadline Make In orderthe to Deadline have your news published in the magazine in a timely fashion, In order to have news published in the thefollowing magazinedates: in a timely fashion, please submit youryour notices to the editor by please submit your notices to the editor by the following dates: Issue Deadline Issue Deadline Jan/Feb October 1 Jan/Feb September 11 Mar/April December Mar/Apr November1 1 May/June February May/June January July/Aug April 1 1 July/Aug March Sep/Oct June 1 1 Sep/Oct May 1 1 Nov/Dec August Nov/Dec July 1 IEEE Engineering in Medicine & Biology Magazine (ISSN 0739-5175) (IEMBDE) is published Engineering bimonthly by Institute Electrical and Electronics Engineers, Inc., IEEE IEEE in The Medicine andofBiology Magazine (ISSN 0739-5175) (IEMBDE) is Headquarters: 3 Park Ave., Floor, of New York, NY Telephone 212 published bimonthly by The17th Institute Electrical and10016-5997. Electronics NY Engineers, Inc.,+1IEEE 419 7900. NY3Telecopier 212 752 New 4929.York, NY Telex: 236-411 (International messages Headquarters: Park Ave., +1 17th Floor, NY 10016-5997. NY Telephone +1 212 only). IEEE Center (for752 orders, address(International changes, Educational 419 7900. NYService Telecopier +1 212 4929.subscriptions, NY Telex: 236-411 messages Activities, Region/Section/Student 445 Hoes Lane, Piscataway, 08854. NJ only). IEEE Service Center (for Services): orders, subscriptions, address changes,NJEducational Telephone: +1 732 981 0060. Price/Publication Information: Individual copies: IEEE Activities, Region/Section/Student Services): 445 Hoes Lane, Piscataway, NJ 08854. NJ Members $10.00 (first981 copy only), nonmembers $20.00 per copy. (Note: add $4.00 Telephone: +1 732 0060. Price/Publication Information: Individual copies:postage IEEE and handling charge to copy any order $1.00 to $50.00, orders). $5.00 per Members $20.00 (first only),from nonmembers $60.00 including per copy. prepaid Subscriptions: year (included in Society fee) for each member of the IEEE Engineering in Medicine and Biology Society.Nonmember Member and nonmember subscription available on inrequest. Biology Society. subscription prices available onprices request. Available microAvailable in microfiche and microfilm. Copyright and Reprint Permissions: Abstracting fiche and microfilm. Copyright and Reprint Permissions: Abstracting is permitted with is permitted creditLibraries to the source. Libraries are photocopy limcredit to thewith source. are permitted to permitted photocopytobeyond thebeyond limits the of U.S. its of U.S. Law Copyright Law for use of(1)patrons: (1) those post-1977 articles carry Copyright for private useprivate of patrons: those post-1977 articles that carry that a code at athecode at the bottom of the first page, provided the per-copy fee indicated in the code is bottom of the first page, provided the per-copy fee indicated in the code is paid through paid through the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA(2) 01923 the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA; preUSA; (2) pre-1978 For allreprint, other or copying, reprint,information, or republication 1978 articles withoutarticles fee. Forwithout all otherfee. copying, republication write information, write CopyrightsDepartment, and Permission Publishing Services, to: Copyrights andto: Permission IEEEDepartment, Publishing IEEE Services, 445 Hoes Lane, 445 Hoes Lane, Piscataway, NJ 08854 USA. Copyright © 2003 by the Institute of Piscataway, NJ 08854 USA. Copyright © 2006 by the Institute of Electrical and Electronics Electrical Engineers, Inc. All rights reserved. Printed in U.S.A. Periodicals Engineers,and Inc.Electronics All rights reserved. Printed in U.S.A. Periodicals postage paid at New York, postage at New York, NY and at additional mailing offices. Postmaster: Sendtoaddress NY andpaid at additional mailing offices. Postmaster: Send address changes IEEE changes to IEEE Engineering in Medicine & Biology Hoes Lane, Engineering in Medicine and Biology Magazine, IEEE,Magazine, 445 HoesIEEE, Lane, 445 Piscataway, NJ Piscataway, NJ 08854 USA. 08854 USA.

Printed in USA

MAY/JUNE 2006

From the Editor the bright future of BME John Enderle

he months of May and June are a time of graduation, a time of new beginnings. I wish all of the graduates the best of fortune and a bright future. This issue focuses on humanitarian efforts by biomedical engineers around the world. The guest editor is Dr. Robert Malkin. He is the director of Engineering World Health and a professor of the practice of biomedical engineering (BME) at Duke University in Durham, North Carolina. I know that you will be amazed by the achievements of the biomedical engineers described in this issue. The future for U.S. BME graduates at all levels appears to be bright. At the American Institute for Medical and Biological Engineering (AIMBE)

T

Academic Council on 3 March 2006, Dr. Steven Schreiner presented data on the 2004–2005 placement statistics. A total of 1,343 graduates were part of the survey, separated by level: B.S., M.S., Ph.D., M.D./Ph.D., and postdoctoral. There were 42 BME programs out of 88 programs providing data. Placement for all was above 90%, with B.S. graduates at 90%, M.S. graduates at 96%, and doctoral graduates at 97%. Of the B.S. graduates, 34% continued their education in graduate school, 21% went into medical school, 35% obtained jobs, and 10% were still seeking employment. Of the M.S. graduates, 43% continued their education in graduate school, 53% obtained jobs, and 4% were still seeking employment. Of

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE Editor-in-Chief John Enderle University of Connecticut Editorial Board Hojjat Adeli The Ohio State University Howard I. Bassen Food and Drug Administration Krzysztof J. Cios Univ. of Colorado at Denver and Health Sciences Center Pouran Faghri University of Connecticut Limin Luo Southeast University, Nanjing Jasjit Suri Biomedical Technologies Inc. Eugene Veklerov Lawrence Berkeley Laboratory Associate Editors A Look At Jean-Louis Coatrieux University of Rennes, France Book Reviews Paul King Vanderbilt University

Cellular & Tissue Engineering Maria Papadaki P&G Italian Research Center, Italy Clinical Engineering Stephen L. Grimes GENTECH COMAR Dennis Blick Independent Consultant Emerging Technologies Dorin Panescu St. Jude Medical Faces and Places Andrew Szeto San Diego State University Genomics Harold (Skip) Garner University of Texas Southwestern Medical Ctr. Government Affairs Luis Kun National Defense University Industry Affairs Semahat Demir National Science Foundation Issues in Ethics John Fielder Villanova University

International News John Webster University of Wisconsin, Madison Patents Maurice M. Klee Fairfield, CT Point of View Gail Baura CardioDynamics San Diego, CA Regulatory Issues Robert Munzner DoctorDevice.com Grace Bartoo Instrumentation for Science and Medicine Retrospectroscope L.A. Geddes Purdue University Senior Design Jay Goldberg Marquette University Society News Jorge Monzon Universidad Nacional del Nordeste Student Activities Jennifer Flexman University of Washington

the Ph.D. graduates, 50% continued their education, 47% obtained jobs, and 3% were still seeking employment. The average starting salary for the B.S. graduate was US$50,400. I would like to encourage IEEE Engineering in Medicine and Biology Society members outside of the United States to share their data on placement of BME graduates. I will post the data as it is received. The AIMBE meeting had a number of interesting themes: Keeping America Competitive, Advancing Women Bioengineers, Fostering Innovation and Accelerating the Commercialization of Inventions, and Biomarkers. While all of the topics were interesting, the one that captured my attention was the role of measure-

IEEE PERIODICALS MAGAZINES DEPARTMENT

Managing Editor Desirée de Myer Art Director Janet Dudar Asst. Art Director Gail A. Schnitzer Business Development Manager Susan Schneiderman +1 732 562 3946 [email protected] Fax: +1 732 981 1855 Senior Advertising Production Coordinator Cathline Tanis Production Director Robert Smrek Editorial Director Dawn Melley Staff Director, Publishing Operations Fran Zappulla

Editorial Correspondence: Address to John D. Enderle, Program Director for Biomedical Engineering, University of Connecticut, Room 223 B, 260 Glenbrook Road, U-2157, Storrs, CT 06269-2157 USA. Voice: +1 860 486 5521. Fax: +1 860 486 2500. E-mail: [email protected]. Indexed in: Current Contents (Clinical Practice), Engineering Index (Bioengineering Abstracts), Inspec, Excerpta Medica, Index Medicus, MEDLINE, RECAL Information Services, and listed in Citation Index. All materials in this publication represent the views of the authors only and not those of the EMBS or IEEE.

4

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

ment technologies (biomarkers, imaging, diagnostics, etc.) in advancing human health. What are the most important measurement technologies and how will they be developed and adopted, including factors and policies that will promote or impede them? Dr. Peter Katona, president of the Whitaker Foundation, gave the Earl Bakken Keynote address, and was the recipient of the Pierre Galletti Award. A remembrance for Swamy Laxminarayan was held during the meeting, with many of Swamy’s family members participating. For more information on the annual AIMBE event, see http://www.aimbe.org/ aimbe/default.jsf.

STEVENS

Institute of Technology

Until the next time, John Enderle

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

5

Society News new chapters, new members Jorge E. Monzon

n this column I would like to mention some issues that illustrate the growth of our Society. Quite frequently, our magazine presents local and regional events organized by our chapters. Local biomedical engineering activities through IEEE Engineering in Medicine and Biology Society (IEEE-EMBS) chapters seem to be a “perpetum mobile,” and we celebrate such continuity as we also welcome the creation of new chapters, which for the last months have arisen in different regions of the world, showing the proactivity of EMBS members everywhere. Hosted by Prof. Jan Bergmans of the Technical University in Eindhoven, the Netherlands, a one-day kick-off meeting of the new IEEE-EMBS Benelux Chapter took place. Bart Vanrumste (Katholieke Universiteit Leuven, Belgium) and Peter Veltink, from the University of Twente (Enschede, the Netherlands) were the organizers of the event. Dr. T.V. Ananthapadmanabha chairs the new EMBS chapter of the Bangalore Section in India. Dr. Nezamoddin N. Kachouie formed a new chapter of the Kitchener Waterloo Section in Canada. A joint chapter— EMBS and the Control Systems Society—has been established in the IEEE Central Coast Section, under the leadership of Dr. Cesar C. Palerm. Another joint chapter—EMBS and Engineering Management—was created by Tom Jobe and colleagues in Oklahoma City, Oklahoma. Our editor-in-chief, Prof. John D. Enderle, and 35 of his students formalized an EMBS student branch chapter at the University of Connecticut. Another student chapter, organized by Hemal Dalai, was created at the Watumull Institute of Electrical Engineering and Computer Technology, in Bombay, India. We also look forward to the success

I

6

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

of those initiatives by Benjamin Mak and Stephen Davies to form a chapter in Toronto, Canada, and by Jan Vrba, who committed himself to start a chapter in the Czech and Slovak Republics. We also welcome the intention of Alexey L’vov to create a chapter in Saratov, Russia. There are some movements in Japan to establish new chapters in Fukuoka and in Sendai under the coordination of Dr. Kenji Sunagawa. Dr. Balakrishnan Santhanaraj, from the city of Kodakara in India, plans to found another EMBS chapter in his country.

Our Society has the highest membership elevated to senior grade. The EMBS Ad Hoc Committee for Chapter Development was created to identify areas that could support EMBS chapter activities and assist local members in the process for creating chapters by providing tools and information that can expedite chapter creation. This committee will be, for 2006, in the hands of Nathalie Gosset (Buenaventura Chapter) and Joaquin Azpiroz Leehan (Mexico Chapter). Among membership development best practices, each year the IEEE promotes its Senior Membership Program. The EMBS Senior Member Initiative was brilliantly coordinated by Barbara Oakley and chapter chairs around the world. Their hard work was highly successful. Sending e-mails to hundreds of potential candidates, distributing forms, contacting references, and answering more than 500 e-mail questions from

people who had received requests or recommendations led to a total of 76 EMBS Senior Member elevations in one quarter. This number places our Society as the one with the highest percentage of its membership elevated to senior grade and quite above all the other Societies. In terms of absolute number of new senior members, EMBS is one of the three most successful IEEE Societies for this program. The other two are • Communications (80 new senior members) • Computer (122 new senior members). This is definitely something to brag about! Just remember that there are 11 IEEE Societies with membership— i.e., potential Senior Members—larger than ours. Although new EMBS senior members belong to all regions, Europe and Asia were the best at growing their senior member base (15 new senior members each), whose loyalty to the Society translates into a higher retention rate. Membership figures (although provisional at the time of this writing) illustrate a positive trend. While a number of other IEEE Societies are losing members, at the end of 2005 our Society had 8,049 members, representing a 2.6% increase from 2004. For the same period, EMBS student membership increased by 11.5%. It must be pointed how, however, that a significant number of members chose to renew IEEE membership but, sadly, did not feel it was valuable to renew their EMBS membership. This suggests that we assess the effectiveness of the advertising campaign for new members and that we reconsider all the services offered to satisfy our members’ needs and as a way of retaining them in EMBS. Our Society continues to grow, as evidenced by those latest reports. No doubt that we all will do our best to keep it that way. MAY/JUNE 2006

Student’s Corner EMBS students surf more than the Web Jennifer Flexman

ou may notice changes to the student Web site—there’s more than just new colors and a new layout. I hope you will find this Web site easy to navigate and useful both for you personally and for your club or chapter. The IEEE Engineering in Medicine and Biology Society (EMBS) student network is growing every day, and we can use this Web site to share ideas, collaborate, and grow professionally. New features on the EMBS Student Web (http://embstudents.org) include: ➤ an online form to submit petitions for establishing student clubs and chapters, and annual renewals ➤ quick links on the home page to take you where you need to go for funding, conference information, summer schools, and more ➤ profiles of events put on by your fellow clubs and chapters to give you new ideas and helpful tips and suggestions ➤ a revamped job site, with links to company databases and resource material

Y

➤ and much, much more!

Please direct any feedback and suggestions to me. Get Ready for EMBC 2006!

Hopefully everyone has submitted their abstracts and is planning to attend the 28th Annual Conference of the EMBS in New York City. Information soon will be provided regarding shared accommodations for students; we will help to organize roommates. In addition, a special session is planned for students: “Students: The Movers and Shakers of EMBS.” Here, you will be able to meet student leaders within EMBS. This session will be a forum for you to ask questions about starting and running a student club or chapter, how to be a leader and motivate others, and what EMBS can do for you to help you plan and fund activities. We want to hear about your experiences, so come ready to share in the conversation. Look for this session in your conference program.

The California Lutheran University EMBS Student Club Mixes Surfing with Tissue Engineering

Thanks to Abigail Corrin for submittng the following to let us know what the California Lutheran University EMBS Student Club has been up to! Our newly established IEEE-EMBS Student Club at California Lutheran University is unlike anything that has ever been held on our campus. The goal of our club is to bring awareness and opportunities to all the science departments on our campus. The group has roots in the Bioengineering Department, but there are also biology, biochemistry, physics, and computer science majors involved. We are networking with the Buenaventura Chapter of the IEEEEMBS to provide mentors and internship opportunities for our students. This club offers a chance for students in the engineering and natural sciences to collaborate with each other about classes, attend conferences, and discuss frustrations and elations regarding undergraduate research. Other than educational excursions, we also attempt to create a close-knit community within our club by doing various activities: having dinner after a conference, taking surfing trips to the beach, and planning an upcoming weekend camping trip. The IEEE-EMBS Student Club brings unique opportunities to California Lutheran University. If you want to share an update about your student club or chapter, please send a description, including pictures, to me ([email protected]). Share Your Opinion

California Lutheran University EMBS students bring biomedical engineering to the beach near Thousand Oaks, California: (back row) Daniel Perkins and mentor Dr. Mike Shawfront; (front row) Abigail Corrin, Erica Freeman, and Mohnish Charan.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

In each “Student’s Corner” column of IEEE Engineering in Medicine and Biology Magazine, I will pose a question, and I want you to respond with your opinion. By encouraging dialogue within our community, I hope to foster MAY/JUNE 2006

7

ing, and in the United Kingdom you can attain Chartered Engineer status. The process varies in other countries. The question is: Do you think that as biomedical engineers we should attain postgraduate licensing for professional designation? What are the advantages/disadvantages of becoming licensed? What kind of licensing exists in your country? Contact me at [email protected] with your responses and comments!

The EMBS student club attends a meeting of the Los Angeles Tissue Engineering Initiative (from left): Erica Freeman, Candace Bragg, Dr. Bill Tawil, Abigail Corrin, Cassandra Hernandez, and mentor Dr. Mike Shaw.

discussion on the broader impact of our field and how it should grow. Being a student is a perfect time to start thinking about these issues because you are the future of the profession. In other types of engineering, where there is frequent interaction with the public, getting the Professional Engineer (PE) designation is commonplace. Being licensed in civil engineering allows you to sign and seal engineering plans and drawings, so it is an essential step in your career. In the United States, getting the PE designation involves writing a general “Fundamentals of Engineering” exam as you

near completing your undergraduate degree, and full licensing is generally granted with four years of acceptable work experience and a pass on the subject-specific “Principles and Practice” exam. However, there is no “Principles and Practice” exam on the subject of biomedical engineering, so students of this discipline can only become licensed in a related field (such as chemical, agricultural, or mechanical engineering) or generally licensed in some states. In Canada, the process is similar, but you can become a Professional Engineer in the discipline of biomedical engineer-

Here are some helpful links: • The National Society of Professional Engineers (USA): http://www. nspe.org/ • The National Council of Examiners for Engineering and Surveying (USA): http://www.ncees.org/ • The Canadian Council of Professional Engineers: http://www.ccpe.ca/ • Engineering Council (United Kingdom): http://www.engc.org.uk/ Jennifer Flexman is currently studying at the University of Washington, Department of Bioengineering (Image Computing and Systems Laboratory/ Neuroimaging and Biotechnology Laboratory) towards a Ph.D. in bioengineering. She graduated with a B.Eng. in electrical engineering from McGill University in 2000 and worked as a wireless test engineer for two years.

Book Reviews

Paul King

Biomedical Imaging Principles and Applications in Engineering Series

Karen M. Mudry, Robert Plonsey, Joseph D. Bronzino (Editors), CRC Press, 2003. ISBN: 0-8493-1810-6, 360 pages, US$109.95. This textbook is a compilation of 16 chapters written by individuals or teams of experts in the fields of medicine, medical instrumentation, and engineering. The chapters were selected from the most relevant sections of the second edition of The Biomedical Engineering Handbook (1999), which was also published by CRC Press. The book is intended for readers with some technical proficiency in engineering and physiology, although it is not geared as a pedagogical textbook for undergraduate or graduate courses. In particular, the chapters neither contain example problems within the text nor homework/thought questions at the end. Nonetheless, most subsections contain a list of defining terms that aid the reader in focusing on the critical features of the topic. The book is subdivided into two sections, the first being devoted to physiologic systems and the second to medical imaging itself. The “Physiologic Systems” section comprises seven chapters covering the primary human systems: cardiovascular, endocrine, nervous, vision, auditory, gastrointestinal (G-I), and respiratory. Although one would not turn typically to a medical imaging book to reference the background material on these systems, these seven chapters do offer a sufficient overview of each system. The “Respiratory System” chapter is the only one in this section that contains several fundamental mathematical models (for gas partial pressures and pulmonary mechanics). The chapter on the “Gastrointestinal System” IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

provides a thorough examination of the control mechanisms and response activity in the G-I tract with specifics about the electrogastrogram (EGG). On the whole, however, this physiology section does not enhance the material presented in the imaging section and could have been omitted entirely from the book. The “Imaging” section contains a chapter on each of the standard imaging modalities—X ray, computed tomography (CT), magnetic resonance (MR) imaging, nuclear medicine (including SPECT), positron emission tomography (PET), and ultrasound as well as several specialty chapters on MR microscopy, electrical impedance tomography, and virtual reality technology. Each of these chapters offers a solid explanation of the components needed for signal/source generation, image detection, and computer processing. Nearly half of the “X-ray” chapter is devoted to the topic of mammography, with many physiological examples related to the image output. Similarly, about half of the “MR Imaging” chapter contains two special-topics subsections on functional MRI and chemical shift imaging. The chapter on CT includes an historical account of the data-acquisition geometries of CT scanners and many good figures illustrating the hardware of the system; several of these figures lack dimensions needed for clarification of size. The chapter closes with an overview of image reconstruction principles. The “Nuclear Medicine” chapter likewise covers system design and image reconstruction techniques. An extensive table of radionuclides used in biomedicine is included in the chapter on PET. The “Ultrasound” chapter opens with a discussion of transducer design and follows with imaging principles. This

part of the text does contain two computational examples on Doppler flow. A summary of the medical applications of virtual reality technology is offered in the final chapter, including coverage of image fusion, surgical training and planning, and telemedicine. Lastly, most of the imaging sections delve into the areas of image resolution and clinical applications. Although published in 2003, almost all of the imaging chapters contain references that are dated no later than 1994, but the developments in each modality are very thoroughly investigated. Because most of the chapters were written by multiple authors, some of the material occasionally is repeated, and thus the structure of the book is not cohesive. Overall, this book would serve best as a reference tool for medical professionals to offer a basic foundation in biomedical imaging systems. —Diane Muratore Testa Western New England College Creative 3-D Display and Interaction Interfaces

B.G. Blundell and A.J. Schwarz, Wiley Interscience, 2006. ISBN: 0471482714, US$94.95. This textbook provides an historical and technical overview of three-dimensional (3-D) display techniques. To a lesser extent, interaction devices are covered. It can serve as a resource for students and others interested in the technology from these perspectives. The text consists of nine chapters. Each consists of the chapter “body,” followed by a discussion/overview of the chapter, and an “investigations” section, which consists typically of six or so semi-homework questions related to the chapter material. A bibliography and reference section is reserved for the end of the text, so reference-chasing means flipping to the end each time. MAY/JUNE 2006

9

Book Reviews (continued)

Chapter 1 in this construct outlines the remainder of the text and gives a 10,000-ft overview of the field of visualization. To further prepare one for the remainder of the text (and a career in the field), chapter 2 consists of an exquisitely well-done chapter covering the human perception of space (vision); chapter 3 covers haptics and the human sensory system. A historical discussion of 3-D visualization, including the use of a vanishing point and horizon in drawing and painting as well as the construction and use of the camera obscura and lucida, are all covered in the next section. Chapter 5 reminds us of traditional interaction mechanisms used in imaging, such as the lightpen, joystick, and mouse as well as the lesser-known implement the grafacon. The Fitts’ model (law) of the human motor system is introduced and described well in this chapter. Chapters 6–8 comprise the “meat” of this text, describing depiction and interaction opportunities (including stereo pairs to virtual reality devices), Haptic devices (used for medical training, prototyping and design and chemistry), and use of the visual channel (an overview of methods for displays of information, including Pepper’s Ghost.) The final chapter, “Adopting a Creative Approach,” primarily discusses the use and usefulness of devices that might allow a two-handed interaction with the data (image) on display. The text concludes with a brief mention of the Chimenti Drawings, an early circa 1850 stereoscopic image pair, and a discussion of holographic images (appendixes 1 and 2.) This text is interesting and comprehensive. It could use a CD insert for many of the images presented; the gray-scale images in the text are generally too small and of too low a quality to be useful for teaching or study. —Paul H. King Vanderbilt University Misadventures in Health Care— Inside Stories

Marilyn Sue Bogner (Editor), Lawrence Erlbaum Associates, 2004. 10 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

ISBN: 0-8058-3378-1, 264 pages, US$59.95. Misadventures in Health Care, Dr. Bogner’s first book since Error In Medicine (1994), is an excellent first volume in a new book series on patient safety published by Lawrence Erlbaum Associates. The book presents a series of independent case studies of episodes of care that involve medical error. The case studies can be read in any order but generally proceed through the continuum of healthcare and include episodes involving the blood bank (i.e., blood transfusion), shock trauma, surgery, intensive care, and anesthesiology, concluding with episodes involving the hospital discharge process and home care. The episodes highlight errors and near misses caused by failures in human-human interaction (e.g., team communication), overreliance on technology (e.g., electronic alarms and alerts), poorly designed human-technology interfaces (e.g., not designed with the human user in mind), and complexity (e.g., too much technology and data overload). Bogner effectively uses the case study approach to provide readers with rare insight into the complexities of the healthcare system from the perspectives of the care providers. Each chapter or case study concludes with a discussion by the authors about the human factors and circumstances that contributed to the error during the episode of care. The authors are experts in patient safety research, and many are practicing clinicians. This is the greatest strength of the book because it provides the case studies with balance of clinical and technical content. The book has been written to facilitate analysis and discussion, so the authors do not propose solutions to the problems identified in each episode. The book succeeds in making the case that the healthcare system is incredibly complex, often for unacceptable reasons, and that the errors and adverse events patients occasionally experience during their interaction with the system seldom result from simple human error. Misadventures in

Healthcare is highly recommended reading for students and professionals in healthcare, human factors engineering, safety and risk engineering, and anyone interested in learning more about safety in healthcare. —Dan France, Ph.D., MPH Design and Development of Medical Electronic Instrumentation

David Prutchi and Michael Norris, 2004. ISBN: 0-471-67623-3, 461 pages, US$126.50. I know these guys. I mean, not personally, but I understand who they are, and I think I have a pretty good idea as to why they wrote this book. They are two veterans of the medical device industry who have dealt with the pressures of creating devices that work in the real world. I expect they wrote the book because they were tired of dealing with young engineers just out of engineering school who understand the theory of electronic instruments but have no idea how to actually build something that works. Having had encounters with student designs using 1-
resistors and 1-F capacitors, I can feel their pain. Clearly they set about creating a book in which they attempt to give the reader real-world examples with realworld components. The book is part textbook, part grimoire, and part technical biography of problems they have solved. As such, it is chock-a-block with component names, values, vendors, and even pin numbers of circuits. And therein lies the trap. In providing such specificity, the authors tie themselves to a particular technological age, a particular style of design, and a particular audience. And in rushing to provide specifics, they miss the opportunity to explain the general. For example, as early as page 9, they show a buffer circuit with a negative-signal guard ring. The part number of the buffer they would use is listed, the pin numbers of the connections are provided, even the jumper connections on a standard input buffer board are provided, but there is no mention of why you MAY/JUNE 2006

want a negative-driven guard ring or even what a guard ring is. There is actually a lot to like about this book. Any book which has major legal warnings about the danger of the circuits or techniques described every ten pages or so holds a certain appeal. They have flyback power supplies, Jacob’s ladders, and toxic chemicals in here! In addition, there are moments, such as in their switched capacitor discussion, where I said, “Oh so that’s why.” The problem is, I have a hard time identifying any sort of sizable audience for it. Its strength is its specificity, which reduces the applicability for a general audience. There are also some strangely dated references. Their A/D system runs through a parallel port, something which is beginning to vanish from modern computers, their code is in QuickBasic—hardly the standard at most universities now—and there are several references to the original Star Trek, which will completely elude the under-45-year-old crowd. I also expect that the modern student won’t understand an analog graphic equalizer, hex addressing for LPT1, or state machine programming. For the right application, I can see this book being invaluable, and the authors are to be congratulated for staking their explanations to reality. However, at US$128 list price and given the small scientific area covered, I can’t see this as being a general medical instrumentation textbook. Final comments to the publisher: If you are going to agree to print a textbook (especially one that you list at US$128), then you should pay some attention to the production values. The figures in the book look they were printed on a dot-matrix printer; Figure 1.3 is a composite of three figures with different font sizes, font types, and line thicknesses. Figure 1.12 looks like it was printed on a plotter. In figures 1.23 through 1.25 “Biopotential amplifier,” “Oscilloscope,” “Input 1,” and “Input 2” are different font types and sizes. If the argument is “that’s how the authors provided them,” then I would hope the IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

authors are getting the lion’s share of the profits. Lastly, the subtitle of this book is A Practical Perspective of the Design, Construction and Test of Medical Devices. As an engineer, I am hardly one to comment on grammar but shouldn’t that be: and Testing of Medical Devices? R.L. Galloway Virtual Reconstruction: A Primer on Computer-Assisted Paleontology and Biomedicine

Christoph P. Zollikofer and Marcia S. Ponce de León, Wiley, 2005. ISBN: 0471-20207-9, US$89.95. If you are interested in, and need some understandable text to comprehend, the three-dimensional (3-D) reconstruction of past and present human and other anatomy, this is the text for you! This is a very well referenced and structured text, covering the fundamentals of image collection and reconstruction in a number of areas related to paleontology and biomedicine. The text does this through a standard chapter format, with a well-conceived set of case studies sprinkled throughout the text, with several appendixes covering standard linear algebra techniques (and others) related to image manipulation, and with suggested readings and references. A companion Web reference site is available for use with the text. The main text consists of eight chapters, each titled according to the main thrust of the section. Chapter 1 is titled “Virtual Reconstruction” and covers the general topic of scanning real objects (fossils to living humans) in order to create an image that is a virtual reality manipulative of the original object. The end goals of this visualization are typically the generation noninvasive reverse-engineered representation of the original. “Data Representation” is the title of the next chapter, which is truly a primer on data structures, image formats, and vision. Chapter 3, “Data Acquisition,” covers the basics of vision versus cameras, transfer func-

tions, various modalities of CT scanning and the basics of MRI as well as surface scanning techniques. Chapter 4, “Image Data Processing,” is a nice overview of the overall process of image formation and analysis. Special emphasis is placed on various imaging techniques, including windowing, filtering, and various boundary extraction methods. Chapter 5 gives an overview of visualization and interaction and covers shading techniques, image movement and rotation, volume rendering, and the use(s) of trackballs, mice, and the like. “Virtual Fossil Reconstruction” (chapter 6) is a good discussion of physical versus virtual reconstruction. Several great examples (fossilized alligator, Neanderthal skull, etc.) give good visualization and instruction into techniques for reconstruction. Virtual surgery for the reconstruction of someone’s face (post bear attack) and custom implant design show a useful application to medicine. Chapter 7, “From Virtual Reality to Real Virtuality,” covers the basics of several rapid-prototyping techniques and gives examples of its use in fossil reconstruction, surgical planning, and implant developments. The final chapter, “Morphometric Analysis,” discusses the use of mathematical tools in the study and analysis of form(s) and change of form(s) over years of evolution and between similar species. An interesting sidelight is a discussion of the commonality of several cave drawings and analyses of shape changes (for example skull shape) as a function of age. For biomedical engineers interested in a practical and understandable text covering the application of virtual reconstruction techniques to problems in medicine, this is an enlightening textbook. It could well comprise a large part of an upper-level medical imaging course. Reading the text is sure to increase one’s vocabulary and likely also make a minor difference (positive) to one’s IQ! —Paul H. King, Vanderbilt University MAY/JUNE 2006

11

BME Company Profiles CleveMed, Orbital Resesarch, and their spin-offs Semahat S. Demir

n this column I am featuring Cleveland Medical Devices Inc. (CleveMed), Orbital Research, iACTIV, Flocel, CleveMed Neuro Wave, and ComSense Technology by presenting an electronic interview I had recently with Robert Schmidt, chairman and founder of these companies.

I

S.S. Demir: Would you please introduce CleveMed and your other companies to us and also the engineering and biosciences that your companies are involved in? Please give us a short history of the companies and how you founded them all. Our readers would be interested in knowing whom your companies employ and what kind of engineers. R. Schmidt: I started Cleveland Medical Devices Inc. (CleveMed) and Orbital Research Inc. in December 1990 and incorporated them the next spring. The intent was to make Robert Schmidt medical devices through CleveMed and aerospace devices and controls through Orbital Research. However, what I found once I started the corporations is that they are more than just a legal person; they develop their own personality and grow into their own entity, despite what the founder initially envisions. As the chief executive officer, you can force the company to follow a predetermined business plan path, but you will stunt its growth by not letting it flourish as the market allows. Consequently, CleveMed is now working for the military developing new wireless systems, and Orbital Research is developing new medical devices. The other companies are spin-offs of these two “mother” companies. 12 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

S.S. Demir: What are your annual sales? Are your companies publicly traded? If yes, what are the annual net incomes? What is your market capitalization? R. Schmidt: Our annual sales of all the companies are about US$10 million. The companies are not publicly traded. They are owned by their employees. S.S. Demir: Where are your companies’ efforts concentrated in bioengineering and biosciences? R. Schmidt: CleveMed’s primary mission is to develop, manufacture, and market wireless monitors and rehabilitation devices to lower total healthcare costs and allow patients to leave the hospital sooner. Our four major focus areas are: emergency brain monitoring, sleep disorders, pressure ulcer management, and movement disorders. Our initial major products are wireless electrophysiology monitors, which are primarily focused on the brain, looking at things like sleep and seizures. Orbital Research has developed a microelectromechanical systems (MEMS) microvalve. These tiny silicon valves are being used to make a Braille array (a computer monitor for the blind). This technology has been licensed to a spin-off, iACTIV Corp. The microvalves can also be used to actuate the microflow effectors on aircraft and missiles. Another MEMS device that we have developed and patented is the Dry Electrode. These micromachined electrodes are used for patient monitoring, and they eliminate the need for pastes or gels, which can dry out over time. S.S. Demir: You founded seven companies: Cleveland Medical Devices Inc. (CleveMed), Orbital, iACTIV,

Flocel, CleveMed NeuroWave, RadioStorm, and ComSense Technology. Please inform us of the entrepreneurship that you have had and the translational activities that your companies had to go through. R. Schmidt: Starting a business from scratch is extremely difficult, particularly without a large source of external funding. When I teach entrepreneurship, I always tell people “never quit a good job to start a company.” I went from a six-figure income to a four-figure income, and that has a drastic effect on your lifestyle. For the first four years, I hardly received any pay, and for the next four years, it wasn’t much better. Most students and professors work very hard for a number of years to get a lab prototype. If they carry their project further, they get to the engineering prototype stage, where their device has the form, fit, and function of the actual product. That means it is the right size and operates pretty much like the final device. Those inventors usually have no concept of the fact that once they have that engineering prototype, they are at best 5% of the way toward commercialization. The next step, and this is a step usually best done by small businesses, is to develop a manufacturing prototype. Now the device has an industrial design that is ready to be made and have customer acceptance; its case and electronics don’t break if you drop it; the software has been tested, not only by in-house engineers but by an actual customer; and it can be made on a production line, without needing individual tweaking. It passes the “ility” tests: reproducibility, reliability, manufacturability, maintainability, reparability, as well as safety tests, and is ready to go through regulatory MAY/JUNE 2006

Fig. 1. The CleveLabs Kit with the BioRadio 150.

Fig. 2. A typical CleveLabs screen shot from the Gait Lab session.

testing. Then you are 10% of the way to a commercial product. In the medical field, the next step is getting your Food and Drug Administration (FDA) approval to market, usually a 510(k) for most devices. That testing and the approval process will get you to about the 25% mark. After that is getting your CMS (Center for Medicare and Medicaid Services) approval for reimbursement. That can be very expensive, and sometimes insurmountable, for new medical devices. Once you have finished that, your new medical device may be half way to commercialization. That last 50% is the market entry piece. It requires getting at least one medical champion and preferably several. These M.D.s are critical. The training of physicians usually takes very smart, creative people and then spends about a decade making sure they stay within the box of “standard of care.” Physicians are trained that if they go outside those bounds, they will put both the hospital and their medical practice at risk. Thus, it takes a brave doctor to perform the medical research and then publish a new way of practicing medicine. From the company’s standpoint, they need not only a brave physician but one who is recognized as a leader in the field. Nothing can be more discouraging than having the data showing that you IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

in fact have a “better mousetrap” but no one cares. Getting people to care not only takes intellectual horsepower but also the funding to get the story out so that other physicians have the opportunity to listen. The model we use is the Thomas Edison Invention Factory model. We have funded a large part of our research using the SBIR program. However, SBIR does not pay for marketing. Since the marketing costs are significantly larger than the development cost, we have to determine how we will fund the marketing of the new product. If we can do it internally, we do so as that gives our current owners the greatest return. However, some products need more funding. We have formed a number of spin-off companies that have raised, or will be raising, money to commercialize the technologies developed by the mother companies. One such example is ComSense Technology Inc. A pressure sensor was initially created at Case Western Reserve University and then further developed by Orbital Research. This pressure sensor has been spun off into a new company called ComSense Technology Inc. This high-temperature pressure sensor is being tested on diesel engines and has the potential for saving up to 5% of transportation fuels, potentially saving up to a bil-

lion barrels of fuel a year worldwide (an amount equal to about 2.5 times the production from Alaskan oil fields). ComSense Technology is in the process of raising additional funding to finalize the product and to start selling it. CleveMed is also in the process of spinning off some of its technologies through emerging companies: CleveMed NeuroWave and RadioStorm. It is intended to license anesthesia neuromonitoring technology and possibly various sleep evaluation and ADHD (attention deficit hyperactivity disorder) diagnosis algorithms to CleveMed NeuroWave and missile tracking and shipboard wireless system evaluation technologies developed for the military to RadioStorm Inc. Flocel is marketing an in vitro blood brain barrier test, a technology we licensed from the Cleveland Clinic Foundation. S.S. Demir: Would you please comment on your companies’ partnerships, collaborations, and alliances? R. Schmidt: We are working with a number of companies in different technology areas; however, most of them are not ready for public release yet. We have started a joint venture in India called Medadim to do sleep testing, initially in Calcutta. MAY/JUNE 2006

13

BME Company Profiles (continued)

S.S. Demir: Would you like to specifically talk about the products of your companies?

Toronto, Arizona State University, Case Western Reserve University, and as far away as Waikato University in New Zealand. We are starting to sell CleveLabs into countries like China, Malaysia, and the Philippines, and the

R. Schmidt: Our Crystal Monitor Model 20-S was designed for the sleep apnea market to provide a deck-ofcards-sized device that allows the patient to be untethered. It is convenient, portable, and FDA approved to market, allowing sleep labs to be set up anywhere: in empty hospital beds, hotels, nursing homes, firehouses, or wherever the user wants to study sleep on a patient. Because it is wireless, an attended study can be performed and the technician can be several rooms away, without the need of running wiring. Another version of the device is the Crystal Monitor 20-E, which is designed for the emergency department (ED) to diagnose nonconvulsive seizures. Our initial testing in an ED, along with other studies, has shown that there are 2–5 million people in the United States that are having nonconvulsive seizures that never get diagnosed. Failure to diagnose and treat nonconvulsive seizures can lead to loss of memory, mental processing problems, and other neurological issues. The Crystal Monitor also has a research and teaching version called the BioRadio 150. It is used for both animal and human research monitorFig. 3. The Crystal Monitor 20-S sleep ing. Uses are as diverse as studying apnea monitor. elderly in their homes by Catholic University in Washington, D.C., to monitoring elephants in Thailand and orcas and dolphins in Siberia. The BioRadio is combined with a laboratory course and is sold around the world as CleveLabs. Because it is used to teach engineering basics, basic electrophysiology, advanced electrophysiology, and clinical applications, it has rapidly become the standard to teach electrophysiology. It has been adopted by universities like the University of Southern California, University of Fig. 4. The Flocel blood brain barrier tester. 14 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

course has also been translated into Spanish. We have also developed a high school course for secondary schools to provide an advantage to their technically minded students. A much smaller version of the BioRadio is sold as the RatPaak. It is approximately the size of a U.S. quarter dollar and is worn in a backpack by small animals to monitor their physiological status. CleveMed has also developed other devices and algorithms. They include: the ParkinSense to monitor Parkinson’s symptoms, Pressore Step to monitor walking and standing of diabetic patients, a Heart Patch and arrhythmia detector for cardiac monitoring, and a number of other devices. Orbital Research is developing a number of flight control technologies. It has shown in wind tunnel tests up to Mach 3.0 that it can control a missile with small micro flow effector devices on the nose cones and tails of missiles. This technology has the potential to reduce the size of missile fins that cause drag, enhancing range and payload and also possibly allowing them to hit targets behind the plane (rear hemisphere engagement) and to provide last-second corrections to be able to more effectively hit moving vehicles. Other flow control technology actively diverts flow through wings and fins using our Reconfigurable Porosity technology. We have also been working closely with the University of Notre Dame to develop a Plasma Wing technology. This Plasma Wing technology is as close to a flying saucer that most people will ever see and can be controlled without any moving parts. Orbital Research has also developed a number of advanced control techniques. One such technique utilizes the neural net of a cockroach’s escape mechanism to provide either collision avoidance or target seeking for missiles, micro-aircraft, sea vehicles, trucks, automobiles, and the like. MAY/JUNE 2006

Valve 1 Valve 2 Valve 3 Valve 7

Valve 4 Valve 5 Valve 6 Valve 8

Card Edge Connector Fig. 5. iACTIV’s MEMS microvalve modules for microfluidics.

Other guidance and control algorithms are developed in near real time using genetic algorithms. This path planning technique randomly selects alternate paths and scores each one based on a set of predetermined criteria such as safety, fuel consumption, or other criteria that may be important to the trip. These algorithms may be used for purposes as varied as planning the flight of a missile or for reprovisioning a ship at sea. A number of other advanced control techniques have also been developed to provide adaptive predictive control and to control underactuated systems (those that are normally thought to be uncontrollable using standard techniques). Both the aerodynamic control devices and the control algorithms have been combined to form a new generation of micro air vehicles. We consider that all of these new controls and vehicles are forming a new industry; we call it aerionics. Orbital Research also works as the distributor of a new heat treating technology that provides an extremely fast quench to provide hardness and toughness to any steel, resulting in properties

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

significantly better than by using current state-of-the-art processes. S.S. Demir: Can you give us an overview of the new areas that your companies will expand into in the near future? R. Schmidt: Our plates are pretty full at the moment. However, the industries we are trying to grow in northern Ohio are neuro, MEMS, and aerionics (our name for our new miniature aviation control systems). Most of our efforts will focus in these three new areas. All three of them are at the bottom of their S curves, so there is lots of room for growth. S.S. Demir: Many of our readers will be interested in hearing about the career and/or job opportunities at your companies. Would you please summarize the career and/or job opportunities? R. Schmidt: CleveMed’s staff is 70% engineers. Most are electrical engineers, including analog, digital, and radio frequency engineers. Software and firmware engineers make up the next

biggest percentage, tied with biomedical engineers, who lead our design teams to develop products that meet human needs. Mechanical engineers and technicians round out the design teams. What we are looking for are more seizure detection engineers that have had electroencephalogram (EEG) experience. At Orbital Research, 75% of our employees are engineers. Most of the Orbital Research engineers are aero/mechanical engineers or materials engineers. We also hire control systems, electrical, and software engineers. S.S. Demir: Please let us know of the Web site from where our readers can get further information about your companies. R. Schmidt: They can see more at: • http://www.CleveMed.com • http://www.Orbitalresearch.com • http://www.iACTIVCorp.com • http://www.ComSenseTech.com • http://www.Flocel.com Two other nonprofits that we support to help grow our industries are: • http://www.NEOBio.org • http://www.OhioMEMS.org.

MAY/JUNE 2006

15

BIOMEDICAL ENGINEERING HUMANITARIANISM

The Humanitarian Efforts of Biomedical Engineers An Overview from the Guest Editor BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

BY ROBERT MALKIN

fter the tsunami hit Southeast Asia, homes, factories, schools, hospitals, and universities were wiped away with the force of thousands of kilograms of rushing water. Immediately, organizations around the world mobilized to help those whose lives were destroyed. Relief organizations sent supplies, professional organizations sent experts, and professional societies (including engineering professional societies) organized relief drives and sponsored their members’ trips to the ravaged area. Some of my colleagues expressed their disappointment that biomedical engineers were seemingly absent in the response to the tsunami, when, arguably, healthcare and healthcare infrastructure were among the most critical needs in the aftermath of the disaster. Taking this one incident as an example of a broad trend, some argue that biomedical engineers are not contributing their share to humanitarian efforts. One of the goals of this special issue is to dispel this myth. Perhaps biomedical engineering (BME) professional societies are not assuming their natural leadership role in delivering humanitarian efforts, but it is simply not true that biomedical engineers are not dedicating a part of their careers towards helping the less advantaged. This special issue of IEEE Engineering in Medicine and Biology Magazine is meant to illustrate by numerous examples the ways in which biomedical engineers are improving conditions throughout the world. Not only do I hope that this series of articles will dispel the myth that BME is not doing its share, but I hope the articles will also inspire an even stronger commitment to humanitarian efforts. And indeed, as scientists and engineers, we have an obligation to commit ourselves to humanitarian work, as Richard Ernst argues in the opening article of this issue. Dr. Ernst won the Nobel Prize in 1991 for his work in the development of magnetic resonance spectroscopy, the cornerstone of modern magnetic resonance imaging (MRI). Like many Nobel Laureates, Dr. Ernst has used his notoriety, in part, to further the cause of humanity, highlighting ways in which we should and must commit ourselves to this cause. There are many ways to fulfill that commitment as a biomedical engineer. Organizations like Engineering World Health (EWH), Volunteer Services Overseas (VSO), and International Aid (IA) describe their need for university faculty

A

16 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

to volunteer their time to teach engineering, science, or math in the developing world. In the case of EWH and IA, this can mean volunteering for as little as few weeks. VSO offers opportunities for up to two years of sustained contribution by a faculty member as an educator in a disadvantaged community. For students of biomedical engineering, there are numerous opportunities to get involved. Senior design projects that focus on the developing world and the domestically disadvantaged are described by Dr. John Enderle, editor-in-chief of this magazine. In some sense, these projects are doubly powerful. They directly contribute to their target audience, but they also can put young biomedical engineers on the path to dedicating a part of their professional lives to humanitarian efforts. Gregory Russell describes a summer program that places students in developing world hospitals, living with developing world families. EWH and Engineers Without Borders (EWB) offer opportunities for engineering students to get involved with designing solutions to problems affecting communities throughout the world. Contaminated water is one the leading causes of death in Africa. Not only students but professionals can get involved in issues such as water purification (such as the project described by Charles Taflin). In these projects, an experienced designer can affect millions with their efforts. If you are interested in volunteering for one of the organizations mentioned here, feel free to contact the authors directly. They look forward to hearing from you. No series of short articles could possibly be a comprehensive list of opportunities for biomedical engineers to carry out humanitarian efforts. But information about BME-specific opportunities is scattered and difficult to find. If you know of an opportunity or an organization that is looking specifically for biomedical engineers and the skills they possess to make a unique contribution to help humanity, please let me know (robert.malkin@ duke.edu). I will accumulate your suggestions and add them to a list that will soon be published. Finally, I hope that these articles will inspire you to consider what you can do to help people suffering from natural disasters, the ravages of war, economic pressures, or other disadvantages. Every biomedical engineer can contribute— including you. (continued on page 19) 0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society: Engineering Revolution in BioMedicine Marriott Marquis Times Square, New York City, New York, USA 30 August – 3 September 2006 Pre-Conference Workshops: 29 – 30 August 2006 Conference Chair Atam P. Dhawan, Ph.D. New Jersey Institute of Technology Conference Co-Chair Metin Akay, Ph.D. Arizona State University Program Chair Andrew F. Laine, Ph.D. Columbia University Program Co-Chair Ki H.Chon, Ph.D. State University of New York, Stony Brook Finance Chair Laura Wolf IEEE-EMBS Publication Chair Andreas Hielscher, Ph.D. Columbia University Local Arrangement Chair Helen Lu, Ph.D. Columbia University Student Activities Chair Jorge Monzon, Ph.D. Universidad Nacional del Nordeste, Argentina Important Dates Four Page Paper Submission: 24 April 2006 Notification of Acceptance: 15 June 2006 Final Program Available: 1 July 2006 Exhibition Booth Reservation: 31 May 2006 Early Conference Registration: 15 July 2006 Hotel Reservation: 1 August 2006

Program Themes Biomedical Signal Processing Biomedical Imaging and Image Processing Bioinformatics and Computational Biology Micro- and Nano- Biotechnologies Bio-sensors, Bio-instrumentation & Wearable Technologies Cellular and Functional Tissue Engineering Pharmaceutical Studies, Drug Delivery and Gene Therapy Biomechanics, Bio-Robotics, and Surgical Planning Cardiovascular and Respiratory Systems Neural and Rehabilitation Engineering, and Neuromuscular Systems Clinical Engineering and Healthcare Information Systems

The 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS) will be held 30 August to 3 September 2006 at the Marriott Marquis Times Square in New York City, USA. The general theme of EMBC 2006 is "Engineering Revolution in BioMedicine" covering the broad spectrum of the medical physics, biological and biomedical sciences, and biomedical and clinical engineering. Specific themes include Molecular Imaging, Neural Engineering, Nano-Technologies for Biomedical Applications, Computational Biology and Bio-Informatics, Technical Innovations in the Pharmaceutical Industry, and BioCounterterrorism and Bio-Defense Technologies. Special symposiums and workshops will be conducted with leaders in the respective fields providing an overview as well as in-depth description of the cutting-edge research technologies. Special emphasis has been given to student activities including opportunities to meet and talk with leaders and pioneers in the biomedical engineering and biological sciences with discussions on exploring career paths. Keynote Speaker: Nora D. Volkow, M.D. Director, National Institute of Drug Abuse (NIDA) August 31, 2006, 8.00 AM-9.00 AM "Imaging the Addicted Human Brain: From Molecules to Behavior" Plenary Speakers: Molecular Imaging: Michael Phelps, Ph.D., Chair, Dept of Molecular & Medical Pharmacology, Director, Crump Institute of Molecular Imaging, UCLA Neural Engineering: John Donoghue, Ph.D., Henry Merritt Wriston Professor and Chairman, Department of Neuroscience, Brown University Nano-Technologies for Biomedical Applications: Jennifer L. West, Ph.D. Isabel C. Cameron Professor and Director, Institute of Biosciences & Bioengineering, Rice University Computational Biology and Bio-Informatics: Joseph M. Jasinski, PhD, Program Director, Life Sciences, IBM Innovations in Pharmaceutical Industry: David Goldenberg, Sc.D., M.D., Chairman, Immunomedics, Inc. Bio-Counterterrorism Technologies: Ernest T. Takafuji, M.D., MPH, Director, Office of Biodefense Research, NIAID, National Institutes of Health Biomechanics: Van Mow, Ph.D., Stanley Dicker Professor of Biomedical Engineering and Orthopedic Bioengineering, Director, Liu Ping Laboratory for Functional Tissue Engineering, Chair, BME, Columbia University Student Paper Competition and Student Design Competition Special Mini-Symposiums on BME Careers, Research Funding, Proposal Writing and Industry-Academia Interactions 18 Technical Workshops and Special Mini-Symposiums

Please visit the EMBC’06 website for further details http://embc2006.njit.edu

BIOMEDICAL ENGINEERING HUMANITARIANISM

Science, Engineering, and Humanity Our Contribution to the Future BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

BY RICHARD R. ERNST

cience, engineering, and humanity. The relationships between these three terms appear to be so obvious that writing about them seems superfluous. We scientists and engineers are convinced that we comprise the primary source of societal prosperity. Innovation is the only path to future happiness! But, is this all that must be said? What about our current trend of destroying the future by exhausting our natural resources, poisoning fertile agricultural soil, and contributing to global warming? What about ruthless economic competition steamrolling weaker cultures, leading to a uniform way of life that leaves behind little deserving of the term culture. Today, cultural diversity is being extinguished as much as biodiversity. And what about our obsession with the development of even more powerful and more inhumane killing devices based on “advanced” scientific principles? Is it really true that we are living in the Robot Age, as was claimed at EXPO 2005 in Aichi, Japan, where the much publicized Robot Project featured floor cleaning robots, garbage collection robots, security robots, guide robots, childcare robots (!!), and next-generation wheelchair robots (http://www-2.expo2005.or.jp/en/robot/ index.html)? Even the opening ceremony was dominated by robots! Why should we have childcare robots when human fertility is decreasing more rapidly than ever before in the so-called civilized countries? The poorer overpopulated countries cannot afford these devices anyway. Perhaps, in the end, all we need are childcare robots that decide, by themselves, when to administer Ritalin to children that become ill-behaved in the hands of other heartless robots! The Swiss producer of Ritalin, Novartis, would be grateful. I hope that these controversial remarks illustrate today’s dilemma. There are many good reasons why we scientists and engineers should be concerned. We the clever inventors of possibly useless devices are also responsible for educating future generations of leaders in politics and economy; they will determine the fate of our offspring. We should implant the proper germs of ideas in their brains. However, we cannot possibly satisfy this vital function by concentrating exclusively on our fascinating research inter-

S

18 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

ests, disregarding the developments in the greater world. We must, at least occasionally, come out of our deep and dark research shafts to appreciate the real problems of today’s world and those that will possibly emerge in the future. Universities are not designed to be just educational institutions for the specialized training of industrial and academic functionaries. Their broader obligations can only be satisfied if they act as radiating cultural centers, directly inspiring and vitalizing the cultural development of our society. Universities should develop innovative and sustainable models for the future development of the global society. We need more interdepartmental discourse at universities and also between science and the humanities for appreciating the most relevant problems of our time and for finding, hopefully, lasting solutions. All members of academic institutions can contribute fruitfully to this discourse. Often, students add the freshest and most unbiased inspirations. Today, as the gaps between societies, cultures, religions, and races widen and deepen at a frightful rate—enhanced by shortsighted and fundamentalist politicians—it is our obligation to actively foster intercultural contacts. Occasions for intercultural encounters are abundant in academia. Let us make lasting contributions to this essential discourse! I can understand why adolescents, full of energy and idealism, decide not to devote their entire lives to the specialized sciences as they are being taught today at traditional universities. Why should an energetic young person work day and night on a particular molecule, elementary particle, or a particular step of a biological mechanism? Research could be much more inspiring when performed in a lively university atmosphere where discussions on humanitarian problem solving are daily activities. Of course, it would not always be possible for research activities and general concerns to coincide. But, sometimes, cross-fertilization might happen, guiding the research in a direction it would not have taken without the lively transdisciplinary discussions of societal issues. François Rabelais (1483–1553) said 500 years ago: “Science sans conscience n’est que ruine de l’âme (Science without conscience is nothing but the ruin of the 0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

As the gaps between societies, cultures, religions, and races widen and deepen at a frightful rate, it is our obligation to actively foster intercultural contacts.

soul).” Indeed, much responsibility rests on the shoulders of academia; it is a long-term responsibility, which political and economic leaders cannot bear because they must solve today’s problems without considering adverse longterm effects. Even dishonesty is accepted, sometimes, from politicians or from profit-oriented managers when their endeavors lead to rapid success. In the academic realm, on the other hand, unconditional honesty forms the very first basic law that should not be violated under any conditions without grave consequences. Our world, sometimes, might appear to be in a hopeless disarray that increases day by day. Nevertheless, we are asked to follow the advice of the great philosopher of science, Karl Popper (1902–1994), who proclaimed in 1993 in Berlin: “Optimism is our duty! We all are responsible for what will come.” Richard R. Ernst was born in Winterthur, Switzerland, in 1933. He enrolled at the Swiss Federal Institute of

Technology in Zurich (ETH-Z), where he received his undergraduate degree in chemical engineering in 1956 and his Ph.D. in 1962, working on high-resolution nuclear magnetic resonance, a field in its infancy at the time. In 1968, Ernst returned to ETH-Z, where he became professor of physical chemistry, from which he retired, reaching the age limit in 1998. His research has dealt with multidimensional nuclear magnetic resonance (NMR) and its practical applications, especially in the medical field. It was for his contribution to NMR spectroscopy that he was awarded the Nobel Prize in Chemistry in 1991. Address for Correspondence: Richard R. Ernst, Laboratorium für Physikalische Chemie, ETH Hönggerberg HCI, 8093 Zürich, Switzerland. Phone: +41 44 632 4368. Fax: +41 44 632 1257. E-mail: [email protected].

Guest Editor (continued from page 16) Robert Malkin is the director of Engineering World Health and a professor of biomedical engineering at Duke University in Durham, North Carolina. Previously, he was the Herbert Herff Professor of Biomedical Engineering at the University of Memphis in Memphis, Tennessee, and the University of Tennessee. Before moving to Tennessee, Malkin was a professor of electrical engineering at the City College of New York and a member of the graduate faculty at the City University of New York and a research associate at Columbia University. Malkin received the B.S. degree in electrical engineering from the University of Michigan in

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

1984, and he received his M.S. and Ph.D. in electrical engineering from Duke University in 1993. Before attending graduate school, he taught English in Thailand, and he worked for EM Microelectronics designing integrated circuits, Cordis Corporation designing pacemakers, and Sarns Inc. designing heart lung machines. He has received numerous awards, including service awards from the Republic of Nicaragua, IEEE-Memphis, EM Microelectronics, and Cordis Corporation; an Outstanding Faculty Research Award from the College of Engineering; an Established Investigator Award from the American Heart Association; and an award for Innovation and Excellence in Undergraduate Education from The President of The City College of New York.

MAY/JUNE 2006

19

BIOMEDICAL ENGINEERING HUMANITARIANISM

The Science of Volunteering A Big Break for Your Career BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

BY PETER CREANE

or many people, taking a sabbatical means time off from their careers. Yet for many others, the thought of taking time off to volunteer overseas is appealing for altruistic reasons, as something they’d like to do as a way of giving back. In recent years there has been an increase in demand from developing countries for skilled professional volunteers. This shift in the delivery of development has led to a fresh look at the personal and professional benefits of volunteering overseas.

F

What Is VSO?

VSO (Voluntary Service Overseas) is an international development charity that works through volunteers. Volunteers live and work with local communities, sharing their skills and knowledge. Since 1958, VSO has sent over 30,000 volunteers to work in Africa, Asia, the Caribbean, and Pacific regions in response to requests from overseas partners, many of them requiring a background in either biology or biomedical engineering. “Opportunities are available for people with specific biology or biomedical engineering skills for medical, technical, and teaching posts around the world. These are highly diverse positions and at any one time we may be looking for a biomedical engineer, laboratory technician, or a biology lecturer,” says Abigail Fulbrook, a placement co-coordinator at VSO. “We look for applicants who can be flexible and can adapt their skills to the needs of their placement. This is important because many jobs involve an element of training or community work, which wouldn’t normally be part of their regular jobs at home.”

There are also personal gains to be had from a career break. Many people want to learn about themselves as much as learn new skills. Responding to new professional challenges in a very different environment is a revitalizing and rewarding experience and can reveal capabilities you never knew you had (see Figures 1–5). At the end of their time away, many volunteers return saying they have learned more than they have passed on. Professionals wishing to volunteer with VSO should be prepared to give two years of their time. This may seem like a long time, but VSO believes that working and sharing skills with colleagues is the most sustainable way to manage development. This type of cross-cultural teamwork takes time to develop. Volunteers need time to test their ideas and to adapt them to different circumstances and demands, and things can move very slowly in developing countries. Many volunteers reflect that they only began to have a real impact during their second year overseas, once they had established themselves as respected and valued members of the community. One in five volunteers chooses to stay longer than the initial two years. A comprehensive package is provided to volunteers, including training before departure and after their arrival in-country. Volunteers also receive a modest living allowance, return flights, accommodations provided by the local employer, medical and travel insurance, national insurance (or equivalent) pension contributions, visas and work permits, and grants before, during, and after the placement, with a guaranteed minimum of three weeks of vacation per year. Case Studies

What’s in It for Me?

The view of employers towards career breaks has changed, especially in recent years with more and more people recognizing that sabbaticals can be a part of career development. Research also shows that career breaks are good for developing skills. A 2001 survey by Demos, an independent U.K. think tank, showed that professional volunteering directly enhanced and developed skills such as adaptability, interpersonal skills, problem solving, strategic thinking, handling responsibility, stress management, and self-assurance, all much sought after qualities in the workplace. 20 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Nigeria: Louise Cuff

In 2002, Louise Cuff, 32, took up a VSO placement as a biomedical engineer at the National Eye Hospital in Nigeria. My role at the hospital was to work with the Instruments Engineering Department. The role of the department was to carry out routine maintenance and repairs on all the medical equipment throughout the hospital. When I first arrived, there didn’t really seem like that much work was going on. There wasn’t any routine maintenance, and repairs would be made, if possible, as and when the equipment failed. No 0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

This shift in the delivery of development has led to a fresh look at the benefits of volunteering overseas.

VSO/FIDAL GO

attempt had been made to repair any of the faulty equipment, whose numbers were growing. The reasons behind these problems became clear when I realized that there were no spares, no training, no test equipment, and no information about the equipment. One of the first things I set about doing with my new team was to try and collect as much information as possible about the existing equipment, much of which did not have service manuals and, where those did exist, they were in German. The level of service we had from suppliers was generally very poor. Very few were interested in helping us. One exception was Carl Zeiss, which was excellent and sent us a whole package of documents, including manuals and drawings of various types of equipment. But, overall, it was very disappointing—considering that when the hospital was commissioned, enough state-of-the-art equipment was purchased from European and American companies to fit seven operating theaters, seven diagnostic clinics, five wards, and a pathology unit, which must have equated to several million British pounds worth of business. We had the same problem locating spares for damaged equipment. Much time was spent on the Internet tracking down the suppliers who invariably didn’t respond to our e-mails for quotations for spares. Eventually, the only way to obtain any was when I returned to the [United Kingdom] for a midterm break and contacted the suppliers directly. Once this was done, they were much more responsive than to my e-mails from Nigeria. Despite these initial frustrations, I loved working at the hospital and am really proud of the things I managed to

Fig 1. David Bevan, science curriculum advisor, showing some students how to use a microscope with the only available regular source of light in Zanzabar, Tanzania— the sun!

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

VSO/LIBA TAYLOR

VSO/PIETERNELLA PIETERSE

Fig 2. Bill Thomas, a volunteer electronics instructor, with medical instrumentation students in Sri Lanka.

Fig. 3. Sarah Ingleby at the Discare Wheelchair Centre in Zambia.

MAY/JUNE 2006

21

Many volunteers reflect that they only began to have a real impact during their second year overseas.

electrical equipment from a very unstable power supply. When I arrived, we had very little, but by the time I left and one of my colleagues took over, we had managed to establish a technical library, set up a well-furbished workshop, and helped the department to gain access training, vital resources, and new skills. The people I worked with were so enthusiastic. They really wanted to learn all they could from me, which in itself was a very motivating and rewarding factor. People seemed so happy despite so many hardships. It really was a quite humbling experience.

VSO/LIBA TAYLOR

Uganda: Eric Bridgeland

VSO/JON SPAULL

Fig. 4. Elenor Clomby, a volunteer Physics Teacher, making new friends in Ethiopia.

Fig. 5. VSO shares skills with local colleagues in India, like this woman.

achieve. When I first arrived, the department didn’t even have its own computer. I managed to secure a grant from VSO, which I used to purchase equipment: a computer, a good soldering iron, RCD switches, and a large number of voltage regulators, which are essential in Nigeria to protect

22 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

After more than three years working in Uganda as a science lecturer at the National Teachers College in Kabale, Eric Bridgeland, 64, can agree with Louise Cuff. I was surprised at how quickly I settled in. Within a month it felt like home, the people were so welcoming and warm. I’d considered volunteering when I was a research student, but I ended up getting married and having kids so it wasn’t really realistic. After 30 years lecturing in biology at Huddersfield University, I was coming up to retirement and felt I needed a change. I still felt I had a lot to offer so I thought I’d try and work in a developing country. I’d never traveled before and, to be quite honest, wasn’t sure what to expect when I arrived. My classes were small at the college, and we had very little resources, but the students were enthusiastic; I always felt my work was really valued and very appreciated. I’ve just accepted a second posting to Ghana and will be leaving in September! Conclusions

Taking a different direction in your life does not have to mean taking time out from your career. Working with VSO may not be a holiday, but it is an experience which will challenge and reward all those involved with memories and experiences that will last a lifetime. To find out more visit http://www.vso.org.uk. Address for Correspondence: Peter Creane, 317 Putney Bridge Road, London, SW15 2PN, United Kingdom. E-mail: [email protected]. Peter Creane is the placement coordinator for health at Voluntary Service Overseas.

MAY/JUNE 2006

BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

Enabling Health-Focused Relief and Development

BY BILLY TENINTY

he World Health Organization (WHO) estimates that in some developing countries, up to 50% of the medical equipment is unusable at any given time. In some hospitals, up to 80% of their medical equipment is inoperative and is stored in hallways or patient rooms. This situation results in the neglect of patients and an increased risk of harm to them and to health workers. In addition, time and resources are wasted on the purchase of sophisticated and duplicate biomedical technology equipment, which is underutilized or never used due to the lack of operator training and a qualified maintenance staff. The inexperience of operators and the lack of repair and maintenance capabilities drastically reduce the functioning life of equipment, limiting access to life-saving care for the most vulnerable segments of society. International Aid (IA), a health-focused relief and development organization, has been addressing this problem by training biomedical technicians in developing countries since 1998. Through their medical equipment training (MET) program, 337 trainees in 15 countries have received at least four weeks of training in electronics, general education, and medical equipment troubleshooting. As of 1 June 2005, 162 trainees have completed 24 weeks of training through the MET program with two classes, one in Ghana and another in Honduras, scheduled to graduate in July and August. The training includes three tracks: electronics, general maintenance education, and equipment troubleshooting. Courses are often taught by volunteer electrical engineers and biomedical engineers. Until his retirement, Phil Marcotte, the first Ghana MET volunteer instructor, taught electronics at Ferris State University (Big Rapids, Michigan). Instructors have come from other universities, such as Ohio State Technical College and Purdue University (Fort Wayne, Indiana). However, the MET program has had volunteers from companies such as Siemens and the local expertise of volunteers from the Hospital Maintenance Systems in Davao City, Philippines, and the National Institute of Mental Health and Neuro Sciences in Bangalore, India. When someone volunteers to teach a course for IA, the organization covers the roundtrip airfare and in-country expenses. IA maintains training centers throughout the world. In Ghana, West Africa, IA’s first and most successful international training site, students have come from many countries,

T

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

BIOMEDICAL ENGINEERING HUMANITARIANISM

International Aid’s Medical Equipment Training Program

including the Democratic Republic of Congo, Ethiopia, Ghana, Kenya, Liberia, and others (see Figures 1 and 2). The Honduras MET program began during summer 2001. Three classes have graduated from the Honduras MET, and a class of 18 students is currently enrolled. In order to perpetuate the program, the Honduras program will be turned over to the host school, Universidad Cristiana Evangelica Nuevo Milenio, in 2006. The Kosovo MET program began in 2001. Graduates are maintaining the medical equipment at the University Clinical Center of Kosovo and the regional hospitals in

Fig. 1. Volunteer Don Slattery teaching Ghana MET students to troubleshoot and repair electrosurgical units.

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

23

Time and resources are wasted on equipment that is underutilized or never used due to the lack of operator training and a qualified maintenance staff.

Fig. 2. Ghana MET students completing repairs on an x-ray machine.

Gjakova, Gjilan, Mitrovica, Peje, and Prizren. Over 100 trainees have entered the Philippine’s MET program. In response to the 26 December 2004 earthquake and tsunami in the Indian Ocean, IA is working with the Aceh Provinical Health Office to set up a medical equipment service (MESC) center in Banda Aceh. The MESC will be staffed by volunteer biomedical engineers working side-by-side with Indonesian electromedical technicians. The first task was to conduct an inventory and assessment of the medical equipment in the province, which will serve as a baseline for MESC. The inventory and assessment of medical equipment was completed in June 2005 and covered 28 hospitals in Aceh Province. An Indonesia MET program will begin in 2008. The MET program has had a tremendous impact. “Our medical equipment technician, Ernest Magbitang, who was

24 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

a laboratory assistant at the time he was sent by the foundation for training on medical equipment, started to attend the MET course on 19 November 2001. After finishing the first module, he can already do minor repair works on our microscopes and suction machines,” says Dr. Raquel M. So-Sayo, medical director of Felimon D. Tanchoco Medical Foundation Hospital, a 220-bed tertiary hospital located in the heart of Caloocan City, Philippines. “With the knowledge gained and skills acquired on the succeeding Modules 2–5, he is now able to help the hospital as well as the college with regard to medical equipment maintenance management. With the technical support of our ME technician, the hospital is able to save on expenditures on repair and maintenance of medical equipment.” International Aid’s MET program has made great contributions to the hospitals in the developing world thanks to the volunteer efforts of biomedical engineers. Instructors are needed for courses in general electronics and specific biomedical areas such as clinical laboratory, cardiac, dental, eye, surgical, ultrasound, and x-ray equipment. Billy Teninty has worked with hospitals in developing countries since 1979. He was trained as a biomedical equipment technician in the United States Air Force and received Association for the Advancement of Medical Instrumentation (AAMI) Certified Biomedical Equipment Technician (CBET) certification in 1980. The medical equipment training (MET) program began in 1998 and has provided tools, test equipment, and training to 372 students in 15 countries. Address for Correspondence: Billy Teninty, International Aid, 17011 W. Hickory Street, Spring Lake, MI 49456 USA. Phone: +1 616 846 7490. Fax: +1 616 846 3842. E-mail: [email protected].

MAY/JUNE 2006

BIOMEDICAL ENGINEERING HUMANITARIANISM

Senior Design for Persons with Disabilities Student Projects That Improve Quality of Life BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

BY JOHN D. ENDERLE

his article provides an overview of senior design in the United States and its impact around the world. Within the United States, the National Science Foundation (NSF) Senior Design Projects to Aid Persons with Disabilities program has provided funding since 1988 to thousands of senior design projects that have been completed by students for persons with disabilities. This program combines the academic requirement of a design experience with enhanced educational opportunities for students, and it improves the quality of life for disabled individuals. Also described are two national design competitions hosted by the Rehabilitation Research’s Rehabilitation Engineering Research Center (RERC) on Accessible Medical Instrumentation (AMI) and the National Collegiate Inventors and Innovators Alliance (NCIIA). In addition, there are a number of biomedical engineering (BME) programs in the United States that provide an opportunity for students to design and construct projects for individuals in developing countries.

T

Introduction

As part of the accreditation process for university engineering programs in the United States, students are required to successfully complete a senior design course during their senior year [1]. Many call this the capstone design course. Senior design in biomedical engineering outside the United States is typically not a part of the undergraduate curriculum. This appears to be due to the three-year duration of many B.S. degree BME programs and the research focus of most BME departments. In Europe, for instance, most BME B.S. degree programs are oriented around a three-year program without a capstone course, followed by a two-year M.S. degree program. There are exceptions, such as the Ecole Polytechnique Fédérale de Lausanne, Switzerland, which requires a two-semester design project, and at the National University of Singapore’s bioengineering four-year program, a two-semester design and research project is required in the senior year. Engineering design is a course or series of courses that brings together concepts and principles that students learn in their field of study—it involves the integration and extension of material learned in their major toward a specific project. Most often, the student is exposed to systemwide synthesis IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

and analysis, critique, and evaluation for the first time. Design is the creative process of identifying needs and then devising a product to fill those needs. It is an iterative, decision-making process in which the student deals with compromise and optimally applies previously learned material to meet a stated objective. It is an approach to problem solving for large-scale, complex, and sometimes illdefined systems. Design is different than research. Design considers alternative solutions and ends by selecting the optimal solution with a fixed goal or specifications in mind. Design often results in a commercial product being developed. Research has an open-ended goal and is exploratory, with no set specifications in mind, and does not necessarily result in a product or a service. It has been reported that over 35 million people in the United States have disabling conditions. More than 9 million Americans have significant mental or physical conditions that prevent them from being able to carry out the major activity of their age group (that is, play, attend school, work, or maintain a household). These numbers are rapidly increasing due to advances in medicine that extend life expectancy. Today, the average American spends approximately 12 years of his or her life as a person with disabilities. Besides the enormous suffering experienced by the disabled community, disability imposes an enormous cost to the United States, totaling more than 6.5% of the gross national product (greater than US$170 billion). Aside from the economic cost to the United States due to disabilities, there is the vitally important consequence of the disability to the individual. Every American has either a disability or direct contact with a person with disabilities (that is, a family member or close friend). Disability ranks as America’s greatest health problem in terms of the number of individuals affected and the economic impact. Devices and software to aid persons with disabilities often need custom modification, are prohibitively expensive, or are nonexistent. Much of the disabled community does not have access to custom modification of available devices and other benefits of current technology. Moreover, when available, engineering and support salaries make the cost of any custom modifications beyond the reach of a person with disabilities. 0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

25

This article describes three activities meant to improve the lives of persons with disabilities. The first is the NSF program started in 1988 to build senior design projects for persons with disabilities. The second is the creation of two national senior design competitions initiated by the Rehabilitation Research’s Rehabilitation Engineering Research Center on Accessible

Fig. 1. This dressing chair was designed by Sarah Park and Ronald Lee (Duke University, 2002) for a little girl named Janie. Janie has cerebral palsy. The chair helps her get dressed for school independently.

Fig. 2. This shoulder-steered tricycle was designed by Derek Juang and Irene Tseng (Duke University BME, 2003) for a boy named David. David has thrombocytopenia-absent radius (TAR) syndrome.

26 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Medical Instrumentation and the BME idea sponsored by the NCIIA. The third is by senior design students from U.S. institutions who are building projects for disadvantaged people in developing countries. National Science Foundation Research to Aid Persons with Disabilities

In 1988, the NSF began a program to provide funds for student engineers at universities throughout the United States to construct custom-designed devices and software for disabled individuals [2]. Through the Bioengineering and Research to Aid the Disabled (BRAD) program of the Emerging Engineering Technologies Division of NSF, funds were awarded competitively to 16 universities in 1988 to pay for supplies, equipment, and fabrication costs for the design projects. (In January 1994, the Directorate for Engineering was restructured. This program is now in the Division of Bioengineering and Environmental Systems, Biomedical Engineering & Research Aiding Persons with Disabilities Program.) Funding for this program has continued each year since, with the goal of this NSF program to enhance the educational opportunities for students and improve the quality of life for persons with disabilities. Approximately 15 universities have been involved with this program on a yearly basis since its start. Students and university faculty provide—through their senior design classes—engineering time to design and build the device or software, and the NSF provides funds for supplies, equipment, and fabrication costs for the design projects. Previously completed projects since 1988 are described at http://nsf-pad.bme.uconn.edu. Figures 1 and 2 show projects completed at Duke University. The purpose of the NSF program is threefold. The first purpose is to provide an opportunity for practical and creative problem solving in addressing a well-defined problem to students for meeting the required design component of their study. An outcome of this involvement is that an individual with a disability receives a device that provides a significant improvement in the quality of his or her life at no cost to the disabled individual due to NSF funding of the projects. In many cases, the development of devices and/or software for an individual may lead to applications for others with similar disabilities. Students are also exposed to a unique body of applied information on current technology in the area of rehabilitation design. The second purpose is to motivate students, graduate engineers, and other healthcare professionals to work more actively in rehabilitation, towards an increased technology and knowledge base, to effectively address the needs of the disabled. This goal assumes greater importance with the implementation of the Americans with Disabilities Act of 1990 (ADA). The third purpose is to allow universities an opportunity for a unique service to the local community. The students participating in this program have been singularly rewarded through their activity with persons with disabilities and have justly experienced a unique sense of purpose and pride in their accomplishments. Many of the projects carried out in this program have been highlighted on national radio, local television news programs, CNN and in conference publications, local newspapers, and the ASEE Engineering Education Magazine. MAY/JUNE 2006

Disability ranks as America’s greatest health problem in terms of the number of individuals affected and the economic impact.

Under faculty supervision in the senior design class, students developed specific projects to address the identified needs of particular disabled individuals. Local school districts and hospitals have participated in this effort by referring interested individuals to the program. A single student or a team of students designs each project for a specific disabled person or a group of disabled individuals with a similar need. A positive outcome of this involvement is that the person with disabilities receives a device that provides a significant improvement in quality of life and independence at no cost to him or her. Students are provided an opportunity for practical and real creative problem solving in addressing a well-defined problem; the person with the disability receives the product of that process. There is no financial cost incurred by the recipients participating in the program and, upon completion, the finished project becomes the property of the individual for whom it was designed. Some of the projects are custom modifications of existing devices, modifications that would be prohibitively expensive to the disabled individual were it not for the student engineer and this NSF program. Other projects are unique one-ofa-kind devices wholly designed and constructed by the student for the disabled individual. University of Connecticut and Ohio University

In the University of Connecticut’s (UConn) BME program, senior design provides an opportunity for practical and creative problem solving. Particular focus in the design experience is placed on the creation of a “commercial product.” As an outreach program, many of the projects built by UConn students are done for persons with disabilities that reside in an economically depressed Appalachian region of southeast Ohio. The goal is to have students from one of the most prosperous states in the United States building projects that make a difference for some of the poorest in the United States. This experience is comprehensive, reflecting all aspects of the engineering design process and industry practice. Students use the Web to describe and report progress on their project. The senior design homepage is located at: http://www.bme.uconn. edu/bme/ugrad/bmesdi-ii.htm. BME senior design at UConn consists of two required courses, BME Design I and II. Design I is a three-credithour course in which students are introduced to a variety of subjects. These include: working on teams, design process, planning and scheduling (timelines), technical report writing, proposal writing, oral presentations, ethics in design, safety, liability, impact of economic constraints, environmental considerations, FDA, patents, manufacturing, and marketing. Design II is a three-credit-hour course following Design I. This course requires students to implement IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

their design by completing a working model of the final product. Prototype testing of the paper design typically requires modification to meet specifications. Ohio University (OU) is well suited to serve as a partner with the UConn senior design project experiences, not only because the two faculty (Dr. Enderle and Dr. Brooke Hallowell) working on the project have complementary areas of expertise but also because of OU’s facilities and clinical affiliations. The College of Health and Human Services at Ohio University houses six health-related professional schools. The college has, in addition to its own on-campus communication disorders and physical therapy clinics, 22 active off-campus clinical contracts serving people of all ages with a wide variety of disabilities, such that contacts with persons with disabilities in the surrounding Appalachian region are numerous. OU also houses its own college of medicine, with active clinical components in geriatric, family, and specialty medicine, providing additional possible sources for the identification of specific needs of individuals in the region. Once the projects are completed at UConn, they are sent to OU for delivery to the clients. To facilitate working with sponsors, a Web-based approach is used for reporting the progress on projects. The student provides a weekly report on the course homepage. The report structure includes: project identity, work completed during the past week, future work, and status review [3] The global marketplace is one in which engineering and professional teams are working on projects simultaneously at distant sites. It is vitally important that students be exposed to working in local teams and global teams at distant sites before working in industry [4]. Part of this experience has included video conferencing, telephone, e-mail, and the use of the Web in communicating project progress between the sponsor, the student team, and the faculty. For the most part, this has worked well. At first, the use of video conferencing was thought to be the best way to communicate. But the difficulty of arranging video conferencing and the cost of video conferencing has limited its usefulness. Telephone, e-mail, and the Web have been the most successful methods of communication. Each year, the NSF has published a book describing each of the successfully completed projects (for example, see [5]). The overall goal for this publication is that it serves as a catalyst and a source of information for future design projects to aid disabled persons. An indirect goal of the publication is to motivate other biomedical engineering programs to work more actively in rehabilitation design and provide service to their local communities. These books are available online at http://nsf-pad.bme.uconn.edu. MAY/JUNE 2006

27

Students are also exposed to a unique body of applied information on current technology in the area of rehabilitation design.

RERC–AMI Annual Student Design Competition: Universal Design for Accessible Medical Instrumentation

An annual design competition has been sponsored by the RERC on AMI for the past two years based on input from focus groups and interaction with participating universities. The goal of this initiative is to bring attention to accessible medical instrumentation. Student teams accepted into the competition received reimbursements up to US$2,000 for their projects. The first year was a test of concept with only a few projects funded while setting up the judging process. In 2004–2005, the second year of the competition, three design competitions were carried out in the following areas: accessible weight scale, accessible syringe dosing, and an accessible ergometer. These projects were designed for a fictitious group of clients with a variety of disabilities, including paralysis, fragility and weakness, multiple sclerosis, diabetes, poor eyesight and blindness, limited limb function due to stroke, Parkinson’s disease, and hearing impairment. For the competition, each team created a Web site to evaluate the design and to help select the winners of the competition. At a minimum, each Web site contained a final report, detailed photos, and a digital video clip of the project in action. The final report also contained a full description of the project (including detailed drawings and photographs, full engineering analysis of optimal design, and at least one alternative design), the consideration of accessible design principles and how the design addressed the needs of the hypothetical clients, and all expenses to build the prototype and the projected cost to create a manufactured product. For full credit, the project was tested with representative intended users, with feedback used to improve the project. The use of appropriate terminology was stressed when dealing with disability and assistive technologies (see http://www.lsi.ku.edu/lsi/internal/guidelines. html). The Web sites followed Web accessibility guidelines (http://www.w3.org/WAI/). Nineteen teams from 16 universities submitted entries into the competition. The projects were evaluated at the end of the spring 2005 semester by 11 judges from government, industry, and academia. At least four judges evaluated each entry. Six teams received awards in the competition for first, second and third places. The results were posted on the RERC Web site (http://www.rerc-ami.org/rcw-sandbox/projects/ d/2/2/year2/). Prizes for first, second, and third place were awarded at US$1,000, US$750, and US$500, respectively. In addition, each team in the competition was given US$2,000 for fabrication costs, parts, and supplies. For the 2005–2006 academic year, three competitions are being offered based on input from the focus groups: blood glucose monitor, accessible medication dispensing device, and a patient positioning aid. As before, these projects were 28 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

designed for a fictitious group of clients with a variety of disabilities. The structure of the competition will remain the same as before for this competition. BME Idea

The BME idea is a national design competition sponsored by the NCIIA. NCIIA supports design through grants and a national design competition focused on invention, innovation, and entrepreneurship to supplement the educational programs at universities across the United States. The goal is to improve traditional classroom instruction by encouraging the development of new technologies that benefit humans and the environment. The design competition does not restrict the type of project, like the RERC on AMI, but does not automatically provide funds to support fabrication costs, parts, and supplies. A ten-page project description is required to enter the competition, which includes: problem objective statement, final report, working prototype, patents search, regulatory issues, manufacturing costs, market analysis, and business plan. The evaluation criteria include technical, economic, and regulatory feasibility. The first, second, and third prizes are US$10,000, US$2,500, and US$1,000, respectively. Additional details are provided at http://www.nciia.org. Design in Developing Countries

This article has thus far described senior design and opportunities in the United States. While there are a few required BME senior design courses in programs outside the United States, there are a number of BME programs in the United States that provide an opportunity for their students to design projects for individuals outside the United States. (This section is based on a presentation by Dr. Matthew Glucksberg at the BME Educational Summit II held in Virginia in March 2005; visit http://www.whitaker.org/academic/wrapup.html.) In the developing world, healthcare needs are quite different; needs often have simpler solutions than those in the United States, and they also have a huge impact. There are a number of diseases significantly impacting the developing world’s population (e.g., HIV/AIDS in Africa, high infant mortality, effects of landmines, environmental hazards, war, nutrition, etc.). In the developing world, resources available are limited: healthcare spending is low, community services such as transportation, electricity, and communications (telephone, cell phone, Internet, etc.) are not available, local and well-equipped hospitals are located only in the largest cities, and skilled workers and industry do not exist. In addition, expectations are significantly different in developing countries regarding healthcare delivery, education, government, religion, etc. Several BME programs in the United States have exported some of their senior design projects to developing countries, focusing on projects that are easy to use, inexpensive, and MAY/JUNE 2006

maintenance free. At Le Tourneau University, Dr. Roger Gonzalez’s students work on extremity prosthetic devices for amputees in Africa. Dr. Lars Olsen from Marquette University had a team of senior design students working on an airway function tester for garment factory workers in Central America. Dr. Matthew Glucksberg’s senior design students working at the Center for International Rehabilitation with a focus on Africa. Dr. Robert Malkin has an extensive program at Duke University (it is described in the article “Prescription for Success” in this issue [6]). All of these programs serve as a model for how other BME programs can make a difference. Conclusions

This article provides an overview of senior design projects in biomedical engineering programs. The NSF Senior Design Projects to Aid Persons with Disabilities program, started in 1988, provides funding for students in their normal ABET course on senior design and provides funding to create and build projects for persons with disabilities. All projects built in this program have been compiled into an annual publication funded by the NSF. The ultimate goal of the annual publications and projects that were built under this initiative is to assist individuals with disabilities in reaching their maximum potential for enjoyable and productive lives. The program at UConn illustrates how senior design can be moved to other sites with profound needs. Two national design competitions in the United States support invention and innovation in senior design projects. The RERC on AMI hosts three design competitions each year, with projects specially described to support a group of fictitious clients. The projects are selected by focus groups identified by the RERC on AMI that highlights accessibility. The BME idea senior design competition is supported by NCIIA, with a focus on promoting innovation and entrepreneurship. The last area described in this article is the novel use of senior design students to build projects for individuals in developing countries. The projects built in these initiatives are typically easy to use, inexpensive, and maintenance free due to the lack of infrastructure in the client’s home country. All of the projects described here have a huge impact on persons with disabilities, especially those who do not have the resources to pay for these projects.

dinator for biomedical engineering at North Dakota State University (NDSU), Fargo. Enderle joined the National Science Foundation (NSF) as program director for the Biomedical Engineering & Research Aiding Persons with Disabilities Program from January 1994–June 1995. In January 1995, he joined the faculty of the University of Connecticut (UConn) as a professor and the head of the Electrical and Systems Engineering Department. In June 1997, he became the director for the Biomedical Engineering Program at UConn. Enderle is a Fellow of the IEEE, the current editor-inchief of IEEE Engineering in Medicine and Biology Magazine, the 2004 EMBS Service Award Recipient, pastpresident of the IEEE Engineering in Medicine and Biology Society (EMBS), EMBS Conference chair for the 22nd Annual International Conference of the IEEE EMBS and World Congress on Medical Physics and Biomedical Engineering in 2000, a past EMBS vice president for Publications and Technical Activities and vice president for Member and Student Activities, fellow of the American Institute for Medical and Biological Engineering (AIMBE), an ABET program evaluator for bioengineering programs, a member of the Engineering Accreditation Commission, a member of the American Society for Engineering Education and Biomedical Engineering Division Chair for 2005, and a fellow of the Biomedical Engineering Society. Enderle was elected as a member of the Connecticut Academy of Science and Engineering in 2003; its membership is limited to 200 persons. He has also been a teaching fellow at the University of Connecticut since 1998. Enderle is also involved with research to aid persons with disabilities. He is editor of the NSF Book Series NSF Engineering Senior Design Projects to Aid Persons with Disabilities, published annually since 1989. These books described almost 2,000 projects that have been constructed and given to persons with disabilities throughout the United States (see http://nsf-pad.bme.uconn.edu). He is also an author of the book Introduction to Biomedical Engineering, published by Elsevier in 2000 (first edition) and 2005 (second edition). Enderle’s current research interest involves characterizing the neurosensory control of the human visual and auditory system.

Acknowledgments

The work described in this article is funded by the National Science Foundation under grants 0302351 and 0454456 and the U.S. Department of Education’s National Institute on Disability and Rehabilitation Research’s Rehabilitation Engineering Research Center (RERC) on Accessible Medical Instrumentation. National Institutes on Disability and Rehabilitation Research, U.S. Department of Education Grant H133E020729. John D. Enderle received the B.S., M.E., and Ph.D. degrees in biomedical engineering and the M.E. degree in electrical engineering from Rensselaer Polytechnic Institute, Troy, New York, in 1975, 1977, 1980, and 1978, respectively. After completing his Ph.D. studies, he was a senior staff member at PAR Technology Corporation, Rome, New York, from 1979–1981. From 1981–1994, he was a faculty member in the Department of Electrical Engineering and coorIEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Address for Correspondence: John Enderle, University of Connecticut, 260 Glenbrook Road, Storrs, CT 06269-2247 USA. Phone: +1 860 486 5521. Fax: +1 860 486 2500. Email: [email protected]. References [1] J.D. Enderle, J. Gassert, S.M. Blanchard, P. King, D. Beasley, P. Hale Jr., and D. Aldridge, “The ABCs of preparing for ABET,” IEEE Eng. Med. Biol. Mag., vol. 22, no. 4, pp. 122–132, 2003. [2] J.D. Enderle, “An overview on the national science foundation program on senior design projects to aid persons with disabilities,” Int. J. Eng. Educ., vol. 15, no. 4, pp. 288–297, 1999. [3] J.D. Enderle, A.F. Browne, and B. Hallowell, “A WEB based approach in biomedical engineering design education.” Biomed. Sci. Instrumentation, vol. 34, pp. 281–286, 1998. [4] J.D. Enderle, W. Pruehsner, J. Macione, and B. Hallowell, “Distance learning in senior design,” in Proc. 26th IEEE EMBS Annu. Int. Conf., July 2000, THBa204, pp. 23–26. [5] J.D. Enderle and B. Hallowell, Eds., National Science Foundation 2003 Engineering Senior Design Projects to Aid Persons with Disabilities, Mansfield Center, CT: Creative Learning Press 2005, p. 405 [Online]. Available: http://nsfpad.bme.uconn.edu [6] G. Russell, “Prescription for success,” IEEE Eng. Med. Biol. Mag., vol. 25, no. 3, pp. 30–31, 2006.

MAY/JUNE 2006

29

BIOMEDICAL ENGINEERING HUMANITARIANISM

Prescription for Success

BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

One Program Corrects the Problems of Hospitals in the Developing World

BY GREG RUSSELL

he first sign that something might be awry at the Children’s Hospital in Managua, Nicaragua, are the sheep grazing in the medical equipment repair area. A better clue, though, would be the fire that unexpectedly erupts above a patient on an operating room table. “A lot of people would be shocked if they saw some of the things that go on in this and other hospitals in third world countries,” Duke University professor Dr. Robert Malkin says. “Some of the occurrences are totally unbelievable.” Malkin was so appalled by such conditions that he created a new institute that focuses on correcting problems at hospitals in economically depressed countries. The program, Engineering World Health (EWH), sends biomedical engineering students to hospitals in underdeveloped countries to repair and install donated medical equipment (Figures 1 and 2). The result has been better care facilities and, in some instances, saved lives. “When we see the people come from the EWH, we are happy because we know improvements are on the way for our hospital,” says Dr. Enrique Alvarado, director of the

Children’s Hospital (Hospital Infantil Manuel de Jesus Rivera) in Managua. “Their work makes our hospital function better and in the process, has helped play a role in saving lives.” Children whose existence relied heavily on antiquated or nonworking medical equipment now have a new life because of support monitors and other equipment installed and repaired by EWH students. Surgeons are better able to operate on patients because of upgraded medical equipment supplied by the Duke-EWH Summer Institute. Malkin says the program is already showing results. He says that Jessica, a two-year-old patient at the Children’s Hospital in Managua, might not be alive today if not for Engineering World Health. Suffering from totally anomalous pulmonary venous drainage, a congenital disorder that is fatal if left untreated, Jessica was completely dependent on the program’s donated, refurbished equipment during her stay at the hospital. “Thanks to our efforts, every station in the intensive care unit of the Children’s Hospital now has a monitoring station,” says Malkin. The program doesn’t just benefit patients either. Those who go through the summer institute receive a unique, hands-on educational experience. “The students get extensive clinical experience, language, and technical training,” says Malkin. “The opportunity to spend a month in a foreign country—especially a third world country—can be a life-changing experience.” The institute is open to engineering, physics, and chemistry majors from any university. “This is an opportunity for me to apply my technical engineering background in an environment that benefits underprivileged children,” says Fig. 1. Students in the Duke-EWH summer program 2005 in Central America about to student Nicolle Kramer. “It makes me embark on a team-building exercise swinging through the trees of the rain forest on aware of the importance of things that zip-lines. After their training, these students worked in poor hospitals in Nicaragua and I would otherwise take for granted.” El Salvador. Malkin says a major problem at many

T

30 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Malkin was so appalled by such conditions, he created a new institute that focuses on correcting problems at hospitals in economically depressed countries.

third world hospitals is not fixing a broken part but having the ability to pay for a part. “One government donated an intensive care unit to the hospital in Managua, and another country donated ventilators to be used in the unit,” Malkin says. “But the hospital was lacking one simple piece needed to make the ventilator work, so the facility was totally empty and not being used. The piece that was missing costs only five dollars, but to the Nicaraguans, that is a lot of money. They just can’t afford it,” he says. “Our students arrive with the needed part, and it has a huge impact even though it is a relatively small amount of money,” Malkin says. “Now the facility can be used.” Lightbulbs are another example. “There is a certain type of lightbulb needed for surgery, but they can’t afford it, so they use regular light bulbs,” Malkin says. “These bulbs can give off too much heat, sometimes resulting in a fire. The nurses have to quickly cover the patient until the fire is extinguished.” Malkin points out that the students in the program fix a wide range of problems. “We try to provide things as simple as lightbulbs or as complex as monitoring stations,” Malkin points out. “On campus, the Duke students meet each week to work on broken equipment. We fix it, recalibrate it, and get it working. It is then shipped down to the developing world where it is installed.” Some Duke students are also taking a new class, Design for the Developing World, where they are designing new equipment to meet the special needs of the developing world. Malkin first became interested in setting up the program when biomedical engineering students approached him “wanting to make a difference” in the world. Malkin says that it is easy for anyone to become involved. Because of a lack of money in the depressed areas, he says that monetary and equipment donations are important to Engineering World Health. Information on donating can be obtained at the Web site (http://www.ewh.org) or by contacting Robert Malkin at [email protected] “The goal of every teacher is to offer a class that is so insightful and exciting that it can change a student’s life,” says Malkin, who holds degrees from the University of Michigan and Duke University. “This educational experience has that potential. It can open students up to an understanding of the real world and their place in that world,” he says.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 2. Santhi Elayaperumal (University of Minnesota, a senior majoring in biomedical engineering) working on an ECG machine. A few hours later it was in use at a patient’s bedside.

“I am convinced we can make a major impact,” he says. “The best thing about it is that it not only provides a great educational experience, it helps serve people who are in need.” For questions about the Engineering World Health program, correspondence should be sent to Dr. Robert Malkin ([email protected]). Greg Russell is editor of the Creative Services Office in the Division of Marketing and Communications at the University of Memphis, a position he has held since November 2000. Before that, he was a writer in the Creative Services Office from 1991–2000. He received a B.A. in journalism from the University of Memphis, Tennessee, in 1985. Address for Correspondence: Greg Russell, 303 Administration Building, University of Memphis, Memphis, TN 38152 USA. Phone: +1 901 678 3811. E-mail: [email protected].

MAY/JUNE 2006

31

BIOMEDICAL ENGINEERING HUMANITARIANISM

Engineers Without Borders and Their Role in Humanitarian Relief Contributing to a Sustainable World BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

BY CLAES I HELGESSON

he first national Engineers Without Borders (EWB) organization, or Ingénieurs San Frontières (ISF), was founded in France in 1982. The vision of ISF-France was to “promote the implementation of sustainable development through the critical practice of engineering by integrating volunteer engineering students in programs managed by professional nongovernmental organizations (NGOs) in Africa, Latin America, and Africa.” ISF also planned to participate in lobbying and information activities in France on issues like fair trade, international trade rules, water and energy supply, etc. Finally, ISF-France engaged, and still engages, ISF groups in the promotion of ethics among engineering students and of critical debates on ethical issues between professionals. Since the foundation of ISF-France, EWB-affiliated, national organizations have been formed in many countries around the world. As of September 2005, EWB-affiliated organizations are registered in more than 50 countries, many of them in the third world. The total number of individual members now exceeds 50,000. Most of the national EWBs are organized with a central, national administration and local, active chapters at universities around the country. The majority of EWB members are engineering students, although in most of the EWB-affiliated organizations, professional engineers also are members. While there has been limited cooperation between EWBaffiliated organizations in different countries so far, an international network organization, EWB-International, was founded in 2002 (http://www.ewb-international.org). Although it is currently based in the United States in Colorado, an international registration has been filed. A discussion has also been initiated to organize regional cooperation platforms in Asia and Europe. In January 2005, EWB-International created “The Humanitarian Engineers Corps Database” as an answer to the direct need to help communities affected by the December 2004 tsunami in Asia (http://www.ewb-international.org/VolunteerDatabase.htm). The database provides a link between those in need and those who can provide services, technologies, and solutions to eradicate poverty in communities around the world. It will also be used for reconstruction and humanitarian projects around the

T

32 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

world. The database is free and open to individuals interested in registering their skills for the benefit of others on the one hand and to organizations involved in poverty reduction on the other. Database members include professionals (active and retired), students, NGOs, humanitarian organizations, etc. All EWB-affiliated organizations share the same vision: a world where all people have access to basic resources and knowledge to meet their self-identified engineering and economic development needs. EWB members want to contribute to new and ongoing development projects around the world in an effective way and at the same time promote new dimensions of experience for engineering students and practicing engineers. It is our belief that this is a primary path to achieving a sustainable world, without suffering the consequences of engineering projects that are socially, culturally, or economically inappropriate. Many national EWBs also participate in lobbying and information activities on sustainability, international and ethics issues like fair trade, international trade rules, water and energy supply, etc. Many EWBs also lobby for changes in the curricula of universities’ engineering programs to include courses on sustainability, fair trade, ethics, etc. Many EWBs also offer their members courses on issues related to project work in developing countries. Why do we just now find such a large interest in EWB and its activities? Many engineering students (and other students for that matter) and professional engineers are frustrated at being tied up with solving problems connected to people in the wealthy part of the world when, at the same time, they are becoming increasingly aware of the poverty that characterizes a majority of the inhabitants in the third world. Many feel that they want to spend some time and effort utilizing their knowledge and experience to bridge the prosperity gap in the world. For a long time, they have had no way to do this, but when the growing, worldwide EWB organization offers a way to spend some time for this cause without jeopardizing their professional careers, the possibility is welcomed. The mission of all EWB-affiliated organizations is to support disadvantaged communities in improving their living standard, welfare, livelihood, and quality of life through 0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Many engineering students and professional engineers are frustrated at being tied up with solving problems connected to people in the wealthy part of the world.

EWB Projects

the implementation of environmentally and economically sustainable engineering projects, while developing internationally responsible engineering students and professionals. EWB members believe in change that can contribute positively to the communities in which they work, in common action to provide new solutions, and in working to interrupt the cycle of poverty that contributes to terrorism and the rejection of democracy. EWB-affiliated organizations believe that the people in host communities must define their own development project goals and contribute to the accomplishment of projects that can solve their problems, thereby building new skills. EWB members believe in environmentally sustainable projects that are symbiotic with the environment, society, and culture. EWB-affiliated organizations also believe in a university education that will develop a new generation of engineers who will benefit from seeing the many facets of engineering solutions to problems in developing communities and not just the technical skills obtained in their basic education. They also believe in the education of host-community partners. EWB members believe in ethics that require the highest level of integrity. Finally, they believe in partnerships with a broad cadre of institutional, academic, development, and engineering professionals who are willing to assist in capacity building towards a more equitable and sustainable world.

Fig. 1. A three-stone oven is the most common oven used in most third world countries. The smoke permeates the hut, causing lung irritation, asthma, and even lung cancer; the stove is also very energy inefficient. Orongo, Ethiopia.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

CAROLINA HILLER

CAROLINA HILLER

EWB projects are carried out by five to ten students and one or more professional engineers. The group spends some weeks, or even months, on location implementing their projects. In some cases, students can receive college credit for their EWB project. Critical issues when carrying through EWB projects are the local contacts and infrastructure on site. Cooperation with local NGOs—or international NGOs who have a history of activities in the country in question—and local authorities is vital. One important way to build relationships with the local people is by regular visits by EWB groups, where one group continues where the earlier group left off. Another important goal is to cooperate with local vocational schools and universities. We are talking with Rotary International to get support from local Rotary clubs in the form of contacts and introductions to local industry and authorities as well as moral support of different types. Most of the projects carried through by EWBs and their members have a direct or indirect healthcare significance.

Fig. 2. A “solar-furnace” developed by students of Chalmers University of Technology, Gothenburg, Sweden, and tested on site in Orongo, Ethiopia.

MAY/JUNE 2006

33

Many EWB projects around the world have been devoted to building water pipes and drilling wells.

Most of the EWB projects have components related to water supply and renewable energy; both are important issues for health. For example, women in most countries in the third world spend a large part of their day collecting firewood to cook the evening meal. The most common stove is a three-stone oven, i.e., a pot standing on three stones heated by an open fire. The smoke from the fire fills the entire hut, which causes lung irritation, asthma, and, in some cases, lung cancer (Figure 1). Many EWB projects have tried to address this serious healthcare issue. One example is the development of a “sun oven,” which consists of foldable reflector plates (Figure 2). By preheating the food before the final cooking, the use of the open firewood could be minimized. Another example is a fan-powered stove that uses solar cells for power, resulting in a lean combustion with a minimum of smoke (Figure 3). Carrying water containers with fresh water for long distances is a daily burden for many young girls in the devel-

Fig. 3. A fan-powered firewood stove developed by the Foundation for Sustainable Tecnologies, FoST, Kathmandu, Nepal (http://www.fost.home-page.org), and implemented in villages around Nepal. The fan can be driven by a small, rechargeable battery charged by solar cells. The fan makes the stove very energy efficient.

34 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

COURTESY OF EWB-USA

CLAES I HELGESSON

oping world. The time spent on water transport could be used for other purposes, such as education. Furthermore, infected water is hazardous, not only when used for drinking or preparing of food but also when used for cleaning dishes and for personal hygiene. When there is a “surplus” of fresh water available, the use of infected water for any purpose will, hopefully, decrease. Many EWB projects around the world have been devoted to building water pipes and drilling wells. Powering water pumps with solar cells offers an interesting and sustainable solution (Figure 4). A combination of modern, efficient solar cells and energy-efficient water pumps has opened the possibility to implement reliable and maintenance-free water pump installations also on remote sites in the third world.

Fig. 4. Sun-powered water pump, Mauritania. The picture shows the installation work in an existing well engineered by a group from EWB-USA.

MAY/JUNE 2006

It is expected that EWB members worldwide will be responsible for a large part of the international aid work in the coming years.

Making EWB Projects Work

All EWB members offer their time and effort to EWB projects without remuneration, although they are reimbursed for their direct costs. The funds to cover the direct project costs are raised through donations from individuals, organizations, and corporations. In many cases, EWB receives grants from governments or international aid organizations. But why do students and professionals offer their time and efforts without remuneration? Many students feel, after having studied for many years, that they want to be able to apply what they have learned to support development in the less privileged parts of the world. Most of the students also believe in the importance of a sustainable world and want to make a contribution towards this goal. Experience shows that engineers who refer to project work under difficult circumstances such as an EWB project in their curriculum vitaes have a competitive advantage when applying for a job. For some students, the EWB project is part of their master’s thesis project. My personal view is that the number of students carrying out EWB projects as an integral part of their education will increase in the near future. This would, however, require changes in the attitude to such projects from the university administration—and teachers. Conclusions

National organizations of EWB and their projects with international healthcare implications are here to stay. It is expected that EWB members worldwide will be responsible for a large part of the international aid work in the coming years, mostly in the areas of water supply and renewable energy. A deeper cooperation between the national EWBs within the framework of EWB-International will also be valuable for students wanting to participate in projects in the third world. This international cooperation could consist of an exchange of experience, allowing project team

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

members from other countries and support to students from other countries wanting to utilize infrastructure and local contacts at a specific site developed by a national EWB organization. Claes I Helgesson received his M.Sc. in chemical engineering at Chalmers University of Technology, Gothenburg, Sweden, in 1959. During his employment as a researcher at the Research Institute for the Swedish National Defense, he earned his Ph.D. in silicate chemistry (ceramic materials) from Chalmers in 1970. During the late 1970s and 1980s, Helgesson was mainly active in implementing results from scientific research into industry. In the 1990s, Helgesson was appointed director (rector) first of the Swedish School of Mining and Metallurgy, Filipstad, and then of the University of Kristianstad, Sweden. Since retirement, Helgesson has been active as consultant professor to Chalmers University of Technology, the University of Gothenburg, and the University of Boras in mainly environmental issues, among others, in research and education. He has been responsible for the development of an International Master’s program on “Implementation of Sustainable Technology to SMEs” (financed by the European Union’s EuropeAid Program, Asia-Link) with students from many parts of the world, including developing countries. In 2001, Helgesson founded EWB-Sweden (http://www.inug.nu), and in 2002, he cofounded EWB-International.

Address for Correspondence: Claes I Helgesson, Lotsgatan 10, SE-414 58 Goteborg, Sweden. Fax: +46 31 775 38 41. Email: [email protected].

MAY/JUNE 2006

35

BIOMEDICAL ENGINEERING HUMANITARIANISM

A Low-Cost Solution to Rural Water Disinfection

BACKGROUND IMAGE© DIGITAL VISION, TLD., INSET PHOTOS, L. TO R., TOP ROW: VSO/FIDAL GRO, 2ND ROW: VSO/LIBA TAYLOR, 3RD ROW: CAROLINA HILLER, VSO/LIBA TAYLOR, 4TH ROW: VSO PIETERNELLA PIETERSE, VSO/JOHN SPAULL.

The Development of an Effective Chlorinator

BY CHARLES TAFLIN

orbidity and mortality statistics show that childhood diarrhea is a major cause of sickness and death in children five years old and younger in developing countries. Other water-borne diseases such as cholera claim the lives of people of all ages in these areas. Providing safe drinking water benefits these populations more than any other single measure. However, providing safe drinking water presents significant biomedical engineering problems. Simple filtration systems that can reduce water-borne pathogen populations by more than 90% are common in many rural areas, but to be truly effective, they should be supplemented by removing or deactivating biological agents (disinfection). Chlorine, an oxidizing agent, is the disinfectant of choice in most water treatment systems because it is widely available and highly effective. Chlorine can also provide a residual disinfectant that will continue to provide protection against contamination throughout the distribution system. For example, in rural Nicaragua there are hundreds of simple water systems serving rural communities with populations of 200 or more. Most of these systems are gravity fed and commonly consist of a spring capture, plastic piping, and one or more small concrete reservoirs. Disinfection, where practiced, usually consists of chlorine bleach, or another oxidizing agent, dripped into the reservoir through plastic tubing. Yet, many problems with this type of chlorine treatment remain, including poor stability of the chlorine and inconsistent drip rates. In 1998, Taller de Salud Campesina (TASCA) reported systems, such as those described for Nicaragua, consistently test positive for coliform bacteria.

M

➤ It should handle water flow rates between 2–10 gallons per

minute (gal/min) and should deliver a consistent dose of chlorine regardless of water flow rate. CTI selected solid tablets of calcium hypochlorite as the source of the oxidizing agent. After two years of development, the chlorinator design was ready. It is called the CTI 8—because it was the eighth design tested—and is constructed of 3- and 4-in diameter polyvinyl chloride (PVC) water pipe and fittings (see Figure 1). In operation, the incoming flow is directed by an influent baffle toward a slotted tube containing a stack of chlorine tablets, which dissolve in the flow. As the lower tablets

Dir

ec

tio

Given the poor success rates with chlorination in the developing world, Compatible Technology International (CTI) decided to tackle this engineering problem. An initial assessment demonstrated that none of the disinfection systems commercially available at that time were economically or technically suited to the task. CTI began the design of a chlorinator more appropriate for the target populations. Our design criteria were as follows: ➤ The unit should be simple, nonelectrical, field-constructible, and suitable for village-level operation and maintenance. 36 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

no

fF

Development of the CTI Chlorinator

low

Cutaway View

Fig. 1. The CTI 8 low-cost water chlorination system is constructed of 3-in and 4-in diameter polyvinyl chloride (PVC) water pipe and fittings. The incoming flow is directed by an influent baffle toward a slotted tube containing a stack of chlorine tablets, which dissolve in the flow.

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Community acceptance is absolutely essential to the ultimate success of this project.

dissolve, new tablets drop into the flow. A flow-related dose is achieved by an effluent weir plate, which controls the water level in the body of the chlorinator (the higher the flow the deeper the water and the more tablets are exposed to the flow). The chlorinator is installed on a bypass line with valves to control the proportion of the total flow passing through the chlorinator. Field Testing

The initial laboratory testing indicated that the device was ready for field testing. In cooperation with the Operation and Maintenance Division (UNOM) of the Nicaraguan Health Ministry, CTI selected a test site and conducted training sessions on the use of the new disinfection technique. During the three months of the initial field test, the village water committee checked the chlorine residual in three locations each day. The initial field results were very promising, leading to the installation of 30 more chlorinators in Nicaragua and two in Guatemala (see Figure 2). CTI has been very sensitive to community acceptance of their new device. Community acceptance is absolutely essential to the ultimate success of this project. For this project, the villagers build, maintain, and operate their own water systems, increasing their investment in the project. The villagers pay a monthly fee, assessed by their own water committee, to support the system and purchase necessary supplies, including chlorine. Technical support and oversight come from the local offices of the government ministries of water and health. The community measures the success of the devices by the palatability of the water they drink and through observing that their children have fewer incidents of diarrhea. Results and Conclusions

While obtaining high-quality chlorine tablets has been a problem, the program has been very well received by the communities, as evidenced by the fact that UNOM cannot keep up with the requests for the units. Laboratory results show that the village water systems with the chlorinators are free of coliform bacteria. Record keeping in these villages is almost nonexistent. However, anecdotal information indicates that infantile diarrhea is reduced when the chlorinator is installed. In approaching a problem like this, there are several factors that will determine the success or failure of the project. The most important has already been mentioned—community acceptance and involvement. Feedback and follow-up are also important; information from the field will not only help you to

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 2. A CTI 8 chlorinator being installed in rural Guatemala.

refine the design but will also show the community that you have a continuing interest in the project. Finally, a truly critical factor is appropriate technology. The device, whatever it is, needs to be engineered for compatiblity with the technological level of the intended users. It may be considerably more difficult to engineer a simple device than a complex one, and the designer must be willing to spend the extra time and effort to “keep it simple.” Charles Taflin’s career spans 50 years, devoted almost entirely to the field of drinking water supply and treatment. His principal employer was the City of Minneapolis Water Works; his last position was as superintendent of plant operations. He also has done water system engineering and treatment plant design for several engineering firms in the Minneapolis-St. Paul area. He has taught courses in water supply and treatment plant design at the University of Minnesota and is one of the contributing authors of the Handbook of Public Water Systems, Second Edition. Address for Correspondence: Charles Taflin, Hamline University, Box 109, 1536 Hewitt Avenue, St. Paul, MN 55104 USA. E-mail: [email protected].

MAY/JUNE 2006

37

The Evolution of Pacemakers

BY SANDRO A.P. HADDAD, RICHARD P.M. HOUBEN, AND WOUTER A. SERDIJN

round 40% of all human deaths are attributed to cardiovascular diseases. Cardiac pacing has become a therapeutic tool used worldwide with more than 250,000 pacemaker implants every year. The purpose of this article is to detail the significant advances in cardiac pacing systems. Our focus is on the evolution of circuit designs applied in pacemakers. Future pacemaker features and further improvements are also pointed out. Since the first artificial pacemaker was introduced in 1932, much has changed and will continue to change in the future [1]–[3]. The complexity and reliability in modern pacemakers has increased significantly, mainly due to developments in integrated circuit (IC) design. Early pacemakers merely paced the ventricles asynchronously, not having the capability of electrogram sensing. Later devices, called demand mode pacemakers, included a sense amplifier that measured cardiac activity by avoiding competition between paced and intrinsic rhythms. By the introduction of demand pacemakers, the longevity increased since pacing stimuli were only delivered when needed. In 1963, pacemakers were introduced having the capability to synchronize ventricular stimuli to atrial activation. Since that time, clinical, surgical, and technological developments have proceeded at a remarkable rate, providing the highly reliable, extensive therapeutic and diagnostic devices that we know today. Modern pacemaker topologies are extremely sophisticated and include an analog part (comprising the sense amplifier and a pacing output stage) as well as a digital part (consisting of a microcontroller and some memory), implementing diagnostic analysis of sensed electrograms, adaptive rate response, and device programmability. Pacemakers have become smaller and lighter over the years. Early devices weighed more than 180 g, whereas today, devices are available weighing no more than 25 g [4]. This weight reduction has occurred partly due to the development of high-energy-density batteries. Finally, there have been remarkable advances in cardiac lead technology. Novel electrode tip materials and configurations have provided extremely low stimulation thresholds and low polarization properties [5]. In this article, we will concentrate on the evolution of analog circuit designs applied in cardiac pacemakers. First, the electrical operation of the heart is described. The following section treats the history and development of cardiac pacing systems as well as their circuit descriptions. Then, some new features in modern pacemakers are discussed. Finally, the conclusions are presented.

A An Electronics Perspective, from the Hand Crank to Advanced Wavelet Analysis

38 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Cardiac pacing has become a therapeutic tool used worldwide.

The Heart Excitation and Conduction System

The heart is composed of atrial and ventricle muscle that make up the myocardium and specialized fibers that can be subdivided into excitation and conduction fibers. Once electrical activation is initiated, contraction of the muscle follows. An orderly sequence of activation of the cardiac muscle in a regularly timed manner is critical for the optimal functioning of the heart. The excitation and conduction system, responsible for the control of the regular pumping of the heart is presented in Figure 1. It consists of the sinoatrial (SA) node, internodal tracks, Bachmann’s bundle, the atrioventricular (AV) node, the bundle of His, bundle branches, and Purkinje fibers. Cardiac cells are able to depolarize at a rate specific for the cell type. The intrinsic rate of AV-nodal cells is about 50 beats per minute (bpm), whereas Purkinje fibers depolarize at a rate of no more then 40 bpm. During normal sinus rhythm, the heart is controlled by the SA node having the highest intrinsic rate of 60–100 bpm, depending on the hemodynamic demand. The right atrial internodal tracks and Bachmann’s bundle conduct the SA-nodal activation throughout the atria, initiating a coordinated contraction of the atrial walls. Meanwhile, the impulse reaches the AV node, which is the only electrical connection between atria and ventricles. The AV node introduces an effective delay, allowing the contraction of the atria to complete before ventricular contraction is initiated. Due to this delay, an optimal ventricular filling is achieved. Subsequently, the electrical impulse is conducted at a high velocity by the HisPurkinje system comprising the bundle of His, bundle branches, and Purkinje fibers. Once the bundle of His is activated, the impulse splits into the right bundle branch, which leads to the right ventricle and the left bundle branch serving the left ventricle. Both bundle branches terminate in Purkinje fibers. The Purkinje fibers are responsible for spreading the excitation throughout the two ventricles, enabling a coordinated and massive contraction [6].

waves depends on the amount of tissue activated per unit of time as well as the relative speed and direction of cardiac activation. Therefore, the physiological pacemaker potentials, i.e. the SA-nodal potentials, generated by a relative small myocardial mass are not observed on the ECG. The first ECG wave within the cardiac cycle is the P-wave, reflecting atrial depolarization. Conduction of the cardiac impulse proceeds from the atria through a series of specialized cardiac structures (the AV node and the His-Purkinje system) to the ventricles. There is a short relatively isoelectric segment following the P-wave. This is the PQ interval, which is related to the propagation delay (0.2 s) induced by the AV node (Figure 1). Once the large muscle mass of the ventricles is excited, a rapid and large deflection is observed on the surface ECG. Depolarization of the ventricles is represented by the QRS complex or R-wave (Figure 2). Following the QRS complex,

Internodal Tracks

Bundle Branch

SA Node AV Node Bundle of His

Purkinje Fibers

Fig. 1. The cardiac conduction system.

R

Cardiac Signals Surface Electrocardiogram

The electrocardiogram (ECG) is a recording from the body surface of the electrical activity generated by the heart. The ECG was originally observed by Waller in 1899 [7]. In 1903, Einthoven introduced electrophysiological concepts still in use today, including the labeling of the waves characterizing the ECG. He assigned the letters P through U to the waves avoiding conflicts with other physiologic waves studied at that time [7]. Figure 2 depicts a typical ECG signal. ECG signals are typically in the range of ±2 mV and occupy a bandwidth of 0.05–150 Hz. The morphology of the ECG IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Bachmann's Bundle

QRS Complex or R Wave: Ventricular Depolarization

P Wave: Atrial Depolarization

ST Segment T

P

P

Q S

PR Interval 0.2 s

T Wave: Ventricular Repolarization

Fig. 2. Typical electrocardiogram.

MAY/JUNE 2006

39

another isoelectric segment, the ST interval, is observed. The ST interval represents the duration of depolarization after all ventricular cells have been activated, normally between 0.25 s and 0.35 s. After completion of the ST segment, the ventricular cells return to their electrical and mechanical resting state, completing the repolarization phase observed as a lowfrequency signal known as the T-wave. In some individuals, a small peak occurs at the end or after the T-wave and is called the U-wave. Its origin has never been fully established, but it is believed to be a repolarization potential [8]. Intracardiac ECG

An intracardiac ECG (IECG) is a recording of changes in electric potentials of specific cardiac locations measured by electrodes placed within or onto the heart by using cardiac catheters. The IECG can be recorded between one electrode and an indifferent electrode, usually more then 10 cm apart (unipolar electrogram) or between two more proximate electrodes (<15 mm) in contact with the heart (bipolar electrogram). Sensing the intrinsic activity of the heart depends on many factors related to the cardiac source and the electrodetissue interface where complex electrochemical reactions take place. In most situations, it is desirable that the IECG does not contain signals from other, more distant cardiac chambers. Bipolar lead systems are much less sensitive to far-field potentials and electromagnetic interference (EMI) sources obscuring the cardiac signal. Cardiac Diseases—Arrhythmias

Arrhythmias (or dysrhythmias) are due to cardiac problems producing abnormal heart rhythms. In general, arrhythmias reduce hemodynamic performance including situations where the heart’s natural pacemaker develops an abnormal rate or rhythm or when normal conduction pathways are interrupted and a different part of the heart takes over control of the rhythm. An arrhythmia can involve an abnormal rhythm increase (tachycardia: >100 bpm) or decrease (bradycardia: <60 bpm) or may be characterized by an irregular cardiac rhythm, e.g., due to asynchrony of the cardiac chambers. An artificial pacemaker can restore synchrony between the atria and ventricles. The History and Development of Cardiac Pacing Artificial Pacemakers

An artificial pacemaker is a device that delivers a controlled, rhythmic electric stimulus to the heart muscle in order to maintain an effective cardiac rhythm for long periods of time,

Power Source

ensuring effective hemodynamic performance. The indication for implanting a permanent pacemaker and selection of the appropriate mode of operation are mainly based on the type of cardiac disease involved such as failure of impulse formation (sick-sinus syndrome) and/or impulse conduction (AV-block). Functionally, a pacemaker comprises at least three parts: an electrical pulse generator, a power source (battery), and an electrode (lead) system (Figure 3) [9]. Different types of output pulses (e.g., monophasic and biphasic) can be used to stimulate the heart. The output stimulus provided by the pulse generator is the amount of electrical charge transferred during the stimulus (current). For effective pacing, the output pulse should have an appropriate width and sufficient energy to depolarize the myocardial cells close to the electrode. Generally, a pacemaker can provide a stimulus in both chambers of the heart. During AV block, ventricular pacing is required because the seat of disease is in the AV node or His-Purkinje system. However, in case of a sick sinus syndrome, the choice of pacemaker will be one that will stimulate the right atrium. A pacemaker utilizes the energy stored in batteries to stimulate the heart. Pacing is the most significant drain on the pulse generator power source. The battery capacity is commonly measured in units of charge (amperehours). Many factors will affect the longevity of the battery, including primary device settings like pulse amplitude and duration and pacing rate. An ideal pulse generator battery should have a high energy density, low self-discharge rate, and sufficient energy reserve between early signs of depletion and full depletion to allow for safe replacement of the device. The electrical connection between the heart and the implanted pulse generator is provided by an implantable electrode catheter called lead. In an implantable pulse generator system, commonly, two types of lead systems are used. A unipolar lead system has a single isolated conductor with an electrode located at the tip. A bipolar lead has two separate and isolated conductors connecting the two electrodes, i.e., the anode and cathode, usually not more than 12 mm apart. The cathode refers to the electrode serving as the negative pole for delivering the stimulation pulse and the anode to the positive pole. For unipolar pacing-sensing systems, the distance between anode and cathode easily exceeds 10 cm. Its cathode is typically located at the lead tip, whereas the pulse generator housing, usually located in the pectoral region, is used as anode. Several types of bipolar leads exist, including the coaxial lead allowing a diameter in the range of 4–5 F (French = 0.33 mm), which is comparable to state-of-the-art unipolar leads. The sensing behavior of bipolar lead systems outperforms their unipolar counterparts by providing a better signalto-interference ratio. Especially for sensing atrial activation, bipolar electrodes are less sensitive to far-field potentials generated by the ventricles. Moreover, bipolar leads are less sensitive to EMI sources and skeletal muscle potentials. However, owing to their construction, bipolar leads are stiffer and more complex from a mechanical construction point of view. Hyman’s Pacemaker

Pulse Generator

Electrodes

Fig. 3. Basic pacemaker functional block diagram.

40 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

In the early 20th century, many experiments such as drug therapy and electrical cardiac pacing had been conducted for recovery from cardiac arrest. Initial methods employed in electrically stimulating the heart were performed by applying a current that would cause contraction of the muscle tissue of the heart. Albert S. Hyman stated that MAY/JUNE 2006

B"

C

Speed Control

Hand Crank

E A

Winds up

Drives Magneto-Generator

Spring Motor Drives

B'

H

Current

G

F Impulse Control

Interruptor Disc Pulsed Current

D I L

K Heart (Right Atrium) Stimulus

J (a)

Needle Electrode

(b)

Fig. 4. (a) The first artificial pacemaker by Hyman. (b) Block diagram of Hyman’s pacemaker.

the introduced electric impulse serves no other purpose than to provide a controllable irritable point from which a wave of excitation may arise normally and sweep over the heart along its accustomed pathways. Hyman designed the first experimental heart pacemaker in 1932 [10] [Figure 4(a)]. Hyman’s pacemaker was powered by a hand-wound, spring-driven generator that provided 6 min of pacemaking without rewinding. Its operation is as follows: The hand crank (F) winds the spring motor (D), which drives the magneto-generator (A) at a controlled speed (E and H) and causes the interrupter disc (not shown) to rotate. The magnetogenerator supplies current to a surface contact on the interrupter disc. The companion magnet pieces (B
and B

) provide the magnetic flux necessary to generate current in the magneto-generator. Subsequently, the interrupter disc produces a pulsed current at 30, 60, or 120 bpm, regulated by the impulse controller (G), which represents the periodic pacing waveform delivered to the electrode needle (L). The neon lamp (C) is illuminated when a stimulus is interrupted. In Figure 4(b), a block diagram of Hyman’s pacemaker is given [11].

Basically, the cardiac pacemaker includes a blocking oscillator [14], which is a special type of wave generator used to produce a narrow pulse. The blocking oscillator is closely related to the more common two-transistor a-stable circuit, except that it uses only one amplifying device—a transistor. The other is replaced by a pulse transformer, which provides inductive regenerative positive feedback. The transistor of the blocking oscillator is normally cut off between pulses and is only conducting during the time that a pulse is being generated. The operation of a blocking oscillator during a single cycle may be divided into three parts: the turn-on period, the pulse period, and the time interval between adjacent pulses (relaxation period). The turn-on period occurs when the supply voltage Vcc is applied to the circuit, R1 and R2 provide a forward bias current, and transistor Q1 conducts. The current flow through Q1 and the primary (L1) of T1 induces a voltage in the secondary (L2). The positive voltage of L2 is coupled to the base of the transistor through C1. This yields more

Vcc = +7.2V

Dawn of the Modern Era—Implantable Pacemakers

The origin of modern cardiac pacing started when the first pacemaker, developed by Dr. Rune Elmqvist, was used in a patient in 1958 by Dr. Ake Senning [12]. In 1959, the engineer Wilson Greatbatch and the cardiologist W.M. Chardack developed the first fully implantable pacemaker [13]. This device was essentially used to treat patients with complete AV block caused by Stokes-Adams diseases, delivering a single-chamber ventricular pacing. It measured 6 cm in diameter and 1.5-cm thick, and the total weight of the pacemaker was approximately 180 g. The pacemaker circuit delivered 1-ms wide pulses to the electrode, a pulse amplitude of 10 mA and a repetition rate of 60 bpm. The average current drain of the circuit under these conditions was about 12 µA, which, energized by ten mercury-zinc cells, gave a continuous operation life estimated at five years. The schematic of the implanted pacemaker is shown in Figure 5 and consists of a pulse forming (square-pulse) oscillator and an amplifier. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

L1

L2

C2 10µF

T1 C1 4.7µF

R4 50kΩ Q1 Q2

R1 750kΩ

C3

10kΩ

10µF

R5

Heart

R3 10kΩ D1

R2 2000kΩ Vcc = +7.2V Fig. 5. A schematic of the first implanted pacemaker.

MAY/JUNE 2006

41

collector current and, consequently, more current through L1. Sufficient voltage is applied very rapidly to saturate the base of Q1. Once Q1 becomes saturated, the circuit can be defined as a series RL (resistance-inductance) circuit, and the current increase in L1 is determined by the time constant of L1 and the total series resistance. During the pulse period, the voltage across L1 will be approximately constant as long as the current increase through L1 is linear. The pulse width depends mainly on the time constant τC = L/R, where R is the equivalent series resistance. After this time, L1 saturates. C1, which has been charged during the pulse period, will now discharge through R1 and cut off Q1. This causes the collector current to stop, and the voltage across L1 returns to 0, shaping the relaxation period. Transistor Q2 implements the amplifier.

in order to detect intrinsic heart activity and thus avoid this competition, one obtains a demand pacemaker, which provides electrical heart-stimulating impulses only in the absence of natural heartbeats. The other advantage of the demand pacemaker compared to the fixed rate system is that now the battery life of the system is prolonged because it is only activated when pacing stimuli are needed. Berkovits introduced the demand concept, which is the basis of all modern pacemakers, in June 1964. In Figure 6, a suitable block diagram of a demand pacemaker is given. Intracardiac electrodes of conventional demand pacemakers serve two major functions, namely pacing and sensing. Pacing is achieved by the delivery of a short, intense electrical pulse to the myocardial wall where the distal end of the electrode is attached, similar to the early pacing devices. However, the Demand Pacemaker same electrode is used to detect the intrinsic activity of the As was shown in the previous section, the early pacing devices heart (e.g., R-waves in the ventricle). The electrical pulse gensimply delivered a fixed-rate pulse to the ventricle at a preset erator consists of the following components: a sense amplifier frequency, regardless of any spontaneous activity of the heart. circuit, a timing control circuit, and an output driver circuit These pacemakers, called asynchronous or fixed-rate pace(electrical impulse former). makers, compete with the natural heart activity and can someThe schematic of the pulse generator designed by Berkovits times even induce arrhythmias or ventricular fibrillation (VF). is given in Figure 7 [15]. The general function of this circuit is By adding a sensing amplifier to the asynchronous pacemaker to make the timing circuit responsive to cardiac activity. This allows inhibition of the pacing pulse from the pulse generator whenever the heart beats on its own. To achieve such function, the sense amplifier plays a funPower damental role. It is designed to amplify and normalize Source the cardiac signal. Also, the sense amplifier is configured to filter out undesired signals such as P- and Twaves and 50-Hz or 60-Hz interference. The electrical signals picked up by the electrodes are coupled through capacitor Cc1 to the input of the amplifier, comprising Sense Timing Output Electrodes transistors Q1 and Q2. The maximum gain of this Amplifier Control Driver amplifier stage is above 50. Alternating current (AC) signals at the collector of Q2 are coupled through capacitor Cc2 to the bases of both transistors Q3 and Pulse Generator Q4. This implements an absolute value function, since signals of positive polarity turn on Q3 and Q5, and Fig. 6. Basic demand pacemaker functional block diagram. signals of negative polarity turn on Q4.

+7V +7V

+ 1.4 V

Q3 + 4.2 V

R11

R12

CC2

RB1

Q5

RE1

C12

C11 RE2

CC1

Rp

CC3

Q2 Q1

Rt

R51

R4 Q4

R3

Q6

RE5

R6 R52

CE2

CE1

RRD

Ct

+ 4.2 V

Rout

Q7 Q8

CC4 +

S R8

CRD

Heart

Q9 R9

_

Z1 Sense Amplifier

Electrodes Timing Control

Output Driver

Fig. 7. A schematic of the pulse generator of the first demand pacemaker.

42 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

When no signal is fed to the base of transistor Q6, Q6 remains nonconducting and will not affect the charging of capacitor Ct . The timing control circuit, which determines the pulse duration (1 ms) and the repetition rate (72 pulses per minute) of the pulse generator, is made up of transistors Q7 and Q8; capacitor Ct; and resistances Rp , Rt , R8 , and R9 . The pulse duration is determined by the time constant τp = Ct Rp and the rate, mainly by τr = Ct Rt . During the charging period, both transistors are off. As Ct charges, this creates a regenerative turn on of both Q7 and Q8, which is sustained as long as Ct can supply current, a time determined primarily by resistor Rp . During this time, the output transistor Q9 is turned on, causing current to flow in the electrode circuit. The output driver comprises transistor Q9, resistor Rout , and capacitor Cc4 . After 1ms, Ct is discharged; transistors Q7, Q8, and Q9 turn off; and the pulse is terminated. Finally, to avoid damage to the circuit due to high-voltage signals from the electrodes, a zener diode (Z1) was placed between the terminals of the electrode. A variation of this concept is the demand-triggered pacemaker, which stimulates every time it senses intrinsic heart activity, i.e., the stimulus falls directly on the natural QRS.

A bandpass filter with a bandwidth of 20–30 Hz is incorporated in the sense amplifier. Three differentiators (RB1 and Cc1 , RE1 and CE1 , and RE2 and CE2 ) limit the low-frequency response of the detecting circuit to discriminate against the Pand T-waves and any other frequencies well below 20 Hz. Two integrators (RI1 and CI1 ; RI2 and CI2 ) are designed to reduce high-frequency noise components well above 30 Hz. However, these filters are not fully effective in preventing the triggering of Q6 by 50 Hz or 60 Hz signals. For this reason, a rate discrimination circuit (including resistors RE5 and RRD and capacitor CRD ) is provided. The switch S is used only to define the operation mode of the system, being either free-running mode (switch closed) or demand mode (switch opened). In free-running mode, the switch is closed and, therefore, transistor Q6 remains cut off. When the switch is opened, i.e., in the case of a pacemaker required to operate in the demand mode, each pulse transmitted through capacitor Cc3 to the base of transistor Q6 causes the transistor to conduct. Capacitor Ct discharges through the collector-emitter circuit of the transistor. In such a case, the timing cycle is interrupted, and the junction of capacitor Ct and resistor Rp does not increase in potential to the point where transistors Q7 and Q8 are triggered to conduction. After capacitor Ct has discharged through transistor Q6, the transistors turn off. The capacitor then starts charging once again, and the new cycle begins immediately after the occurrence of the last heartbeat.

Dual-Chamber Pacemaker

A dual-chamber pacemaker typically requires two pacing leads: one placed in the right atrium and the other placed in

+7V +7V

R3

Q3 + 4.2 V

+ 1.4 V R11

CC2

R12

R4

Rt

R51

Q4 Q5 RB1

Q2

RE2

CC1 RE1

C12

C11

Q1

Rp

CC3 Q6

RE5

R6 R52

CE2

+ 4.2 V

Rout

Q7

Ct

Ventricular CC4 Electrode

Q8

S R8

CE1

RRD

CRD

Neutral Electrode

Q9 R9

Z1 Sense Amplifier

S FET

Ventricular Timing Control

Da

Output Driver

Rta

Ca Ra

Rpa Q6a

Cta

R6a

Q7a Q8a

Routa Atrial CC4a Electrode

R8a

Za Q9a

R9a Atrial Timing Control

Output Driver

Fig. 8. Schematic of the dual-chamber demand pacemaker.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

43

Since the first artificial pacemaker was introduced in 1932, much has changed and will continue to change.

the right ventricle. A dual-chamber pacemaker monitors (senses) electrical activity in the atrium and/or the ventricle to see if pacing is needed. When pacing is needed, the pacing pulses of the atrium and/or ventricle are timed so that they mimic the heart’s natural way of pumping. Dual-chamber pacemakers were introduced in the 1970s. One of the first descriptions of a dual-chamber pacemaker was given by Berkovits in 1971. Berkovits announced a bifocal (AV sequential) pacer that sensed only in the ventricle but paced both chambers. In the presence of atrial standstill or a sinus-node syndrome plus AV block, the bifocal pacemaker could deliver a stimulus to the atrium and then, after an appropriate interval, to the ventricle. Berkovits improved on his original design given in Figure 8 with a dual-chamber demand pacemaker. A schematic of this design is given in Figure 8 [16]. In accordance with the principles of the demand pacemaker design, a sense amplifier is provided to detect intrinsic ventricular activity. The timing control circuits determine both atrial and ventricular time-out stimulating period. However the atrial-stimulating impulse is generated first, and, after a predetermined time interval (200 ms), the ventricular-stimulating impulse is generated. Three electrodes are provided: a neutral electrode, an electrode for atrial stimulation, and an electrode for ventricular pacing and sensing. The field-effect transistor (FET) switch (S FET) is inserted in the feedback path of the ventricular electrode in order to avoid erroneous detection because of the atrial contraction. The S FET is normally conducting. The negative pulse generated at the atrial electrode is transmitted through the diode Da , charging the capacitor Ca , and turning off the switch. When the atrial-stimulating terminates, Ca discharges through resistor Ra and turns on the switch again. In this manner, the sense amplifier is disabled during each atrial stimulation and for a short interval thereafter.

More sophisticated dual-chamber pacemakers that sense intrinsic activity and pace in both chambers were developed, with their first use in late 1977. Rate-Responsive Pacemaker

The latest innovations include the development of rate-responsive pacemakers in the early 1980s, which could regulate their pacing rate based on the output of a sensor system incorporated in the pacemaker and/or lead. A sensor system consists of a device to measure some relevant parameter from the body (e.g., body motion, respiration rate, pH, and blood pressure) and an algorithm in the pacemaker, which is able to adjust the pacemaker response in accordance with the measured quantity. Modern rate-responsive (also called frequency-response) pacemakers are capable of adapting to a wide range of sensor information relating to the physiological needs and/or the physical activity of the patient. A block diagram of a rate-responsive pacemaker is given in Figure 9. The system is based on a pacemaker having a demand pulse generator, which is sensitive to the measured parameter. Many rate-responsive pacemakers currently implanted are used to alter the ventricular response in singlechamber ventricular systems. However, rate-responsive pacing can also be done with a dual-chamber pacing system. New Features in Modern Pacemakers Detection and Sensing Circuitry

A modern pacemaker consists of a telemetry system, an analog sense amplifier, analog output circuitry, and a microprocessor acting as a controller. Nevertheless, the sense amplifier plays a fundamental role in providing information about the current state of the heart. State-of-the-art implantable pulse generators or cardiac pacemakers include real-time sensing capabilities that are designed to detect and monitor intracardiac signal events (e.g., Rwaves in the ventricle). A sense amplifier and its subControl Algorithm sequent detection circuitry, together called the front-end, derive only a single event (characterized by Logic and Control a binary pulse) and feed this to a microcontroller that Sensors Section decides on the appropriate pacing therapy to be delivered by the stimulator. Over the years, huge effort has been put into the improvement of sense amplifier and detection circuitry. The dynamic range of the atrial and ventricular Sense Timing Output Electrodes electrograms sensed by an endocardial lead typically Amplifier Control Driver lies between 0.5–7 mV and 3–20 mV, respectively. Slew-rates of the signals range 0.1–4 V/s. For the QRS complex, spectral power concentrates in the Demand Pulse Generator band 10–30 Hz. The T-wave is a slower signal component with a reduced amount of power in a band not Fig. 9. A block diagram of a rate-responsive pacemaker.

44 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

exceeding 10 Hz. Amplification of intrinsic cardiac signals requires circuitry that is robust against artifacts generated from noncardiac electromagnetic sources located outside or inside the patient. Introduction of electronic article surveillance (EAS) systems has raised concerns with regard to the possible interaction between emitting field sources and the sense amplifiers of medical implantable devices like pacemakers [17], implantable cardioverter defibrillators, and insertable loop recorders (ILRs) [18]. Other sources of EMI include cellular phones, airport metal detector gates, high-voltage power lines [19], electro-cautery devices, and magnetic resonance imaging (MRI) equipment [20]. The more sensitive atrial-sensing channel of a brady-arrhythmia device is more prone to detection of EMI. Any type of EMI having sufficient amplitude could cause the pacemaker to react in a clinically undesirable way, either inhibiting or triggering stimuli. Fortunately, noise reversion algorithms and circuits mostly provide reliable discrimination between EMI and intrinsic cardiac activity. In Figure 10, a suitable block diagram of a sense amplifier for cardiac signal detection is given [21]. The IC or chip consists of a voltage-to-current (V-I) converter, a bandpass filter, an absolute value circuit, an adaptive threshold circuit, and a comparator circuit. Additionally, an EMI filter is implemented outside the chip for EMI cancellation. This usually is a second-order bandpass filter to suppress dc and signals beyond 1 kHz. The V-I converter is required as the input and output quantities of the EMI filter are voltages, and the applied circuit technique for the remainder of the sense amplifier is inherently current-mode. The bandpass filter is used to specifically select intracardiac signals, in our case being the QRS complex or R-wave, and to minimize the effect of the overlapping myocardial interference signals and low-frequency breathing artifacts. The center frequency of the bandpass filter is located at 25 Hz. The reason to use an absolute value circuit is to be independent from the electrode position in the heart. Accommodation to changes in the average input signal level is realized using the adaptive threshold circuit. At the end of the block schematic, the detection signal (a binary value) is generated depending on a threshold level.

Off-Chip

On-Chip

EMI Filter

Bandpass (Two Biquads) Filter

A test signal is applied to the system to verify the performance and efficiency of the complete sense amplifier according to Figure 10. A typical intracardiac signal measured in the ventricle, shown in Figure 11(a), is applied to the input of the system. The transient response of the circuit is shown in Figure 11(b). The system is clearly able to detect the R-wave, which represents the cardiac event that the circuit was supposed to detect. Morphological Analysis

In pacemakers, one of the challenges is the reduction of unnecessary therapies delivered to the patient when the heart rate dynamics becomes comparable to that of lethal tachyarrhythmias like ventricular tachycardia (VT) or VF. This situation includes supra-VT (SVT) that may occur as a result of atrial fibrillation. As heart rate does not discriminate between lethal tachyarrhythmias like VT/VF and SVT or atrial tachyarrhythmias, the morphology of the QRS complex, or more specifically, the R-wave morphology, can be used for a more accurate discrimination between SVT and VT. In addition, to ensure efficient use of the memory available in an implantable device, the incidence of false positives, erroneously triggering automatic storage, should be minimized. For ILRs, promoting factors include the low-amplitude electrogram signal as a result of the limited vector available for pseudo-ECG measurement and the presence of muscle electromyography (EMG) and mechanical disturbance of the electrode tissue interface. Therefore, signal analysis methods improving discrimination of signals from noise are of great importance. Since, usually, signal and noise components share the same spectral bands, the scope of linear signal processing methods is rather limited. Since the information retrieved by the above front-end circuit is reduced to a single event, morphological attributes of the electrogram are completely suppressed. Recent research and clinical studies report details on how morphological aspects of the electrogram relate to various pathological states of the heart and on how the wavelet transform (WT) can contribute efficiently to analysis.

Absolute Value

V-I Converter

Comparator Adaptive Threshold Circuit

Ι

1

Event Detection

0

6.0m 3.0m 0.0 −3.0m 26

1.0n Lout

Vin

Fig. 10. A block diagram of a sense amplifier.

27

28

29 Time (s) (a)

30

31

32

500p 00 26

27

28

29 30 Time (s) (b)

31

32

Fig. 11. Transient response of the complete sense amplifier circuit: (a) input voltage and (b) output current.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

45

The WT is a merited technique for the analysis of nonstationary signals, like cardiac signals. Signal analysis methods derived from wavelet analysis [22] carry large potential to support a wide range of biomedical signal processing applications, including noise reduction [23], [24], feature recognition [25], and signal compression. Wavelets allow the analysis of the electrogram by focusing on the signal at various levels of detail in analogy with inspection of a sample with a microscope at various levels of magnification. As one can see in Figure 12, at very fine scales (smaller values of Scale a), details of the electrogram, i.e., the QRS complex, are revealed while unimpaired by the overall structure of the signal. At

Coefficients (Absolute Values)

QRS

14,000 12,000 10,000 8,000 6,000 4,000 2,000 1 5 9 13 17 21 25 29 33 37 Scale a 4145 49 53 57 61

100

200

300

400

Fig. 12. Wavelet analysis of an intracardiac signal.

Filtering Stage Wavelet Transform Filter

coarse scale (larger values of the Scale a), the overall structure of the electrogram can be studied while overlooking the details. Note that by this global view, both the QRS complex and T-wave can be detected. Analyzing the structure of the electrogram over multiple scales allows discrimination of electrogram features over all scales from those only seen at fine or coarse scales. Based on such observation, the presence or absence of electrogram features related to proximal or distal electrophysiological phenomena can be discriminated. Then, being a multiscale analysis technique, the WT offers the possibility of selective noise filtering and reliable parameter estimation. An algorithm based on wavelet analysis that compares morphologies of baseline and tachycardia electrograms based on differences between corresponding coefficients of T their WTs has been found highly sensitive for VT detection [26]. Whereas smoothing attenuates spectral components in the stop band of the linear filter used, wavelet denoising attempts to remove noise and retain whatever signal is present in the electrogram. Off-line ECG analysis, like Holter analysis, employs the discrete WT implemented in the digital domain using multi-rate filter banks [27]. In these applications, the WT provides a means to reliably 900 1,000 600 700 800 detect QRS complexes. 500 However, in patient-worn Time external applications (e.g., intelligent Holter devices), it Decision Stage Absolute Value

Sign Function

t−τ 1 ∞ C(τ,a) =  ∫x(t)ψ∗ dt a √ a −∞

Abs

t−τ 1 ∞ a = 22 C(τ,a) =  ∫x(t)ψ∗ dt a √ a −∞

Abs

t−τ 1 ∞ C(τ,a) =  ∫x(t)ψ∗ dt a √ a −∞

Abs

t−τ 1 ∞ C(τ,a) =  ∫x(t)ψ∗ dt a √ a −∞

Abs

∞ a = 25 C(τ,a) = 1 ∫x(t)ψ∗ t − τ dt  a √ a −∞

Abs

Cadiac Signal a = 2

1

a = 23

a = 24

Peak Detector

Peak

Peak Detector

Peak

Peak Detector

Peak

Peak Detector

Peak

Peak Detector

Peak

1 0 1 0 1 0 1 0

D I G I T A L

Event

L O G I C

1 0

Fig. 13. Block diagram of the wavelet system.

46 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

J=4

is not favorable to implement the WT by means of digital sigmaxima of the QRS complex for a specific scale (a = 24 ) of nal processing due to the high power consumption associated the WT indeed have been detected. with analog-to-digital conversion and computation. The obtained results for a typical cardiac signal Therefore, a method for implementing the WT in an ultrademonstrate a good performance in generating the desired low-power analog way by means of dynamic translinear WT and achieving correct QRS complex detection. (DTL) circuits has been developed [28], [29]. The main The Electronics Research Laboratory of Delft University of advantage of DTL circuits with respect to other low-power Technology, together with the University of Maastricht and techniques is the ability to handle a large dynamic range in a Medtronic Bakken Research Center, is currently investigating low-voltage environment. It allows for the implementation of a fully integrated implementation of the analog WT circuit to a large variety of transfer functions described by (possibly be used in pacemakers. In addition, further analysis techniques nonlinear) polynomial differential equations. Moreover, only transistors and capacitors are required to realize 600p 300p these functions. Since, in conventional 0.00 ultralow-power designs, resistors −300p 22.5 23.0 23.5 24.0 24.5 would become too large for on-chip Time (s) integration, their superfluity is a very important advantage. Other advantages (a) are that DTL circuits present a high 1.0n functional density and are theoretically 500p process and temperature independent. 0.0 In Figure 13, a first prototype −500p 22.5 23.0 23.5 24.0 24.5 wavelet-based system has been Time (s) defined [28]. At the input, a wavelet (b) filter is situated that implements an approximation to the first derivative 300p Gaussian WT, a function most widely 100p used for frequency analysis among −100p −300p wavelet functions. The complete filter 24.0 23.0 23.5 24.5 22.5 comprises multiple scales in parallel Time (s) in order to compute the WT in real 500p 200p time. As mentioned previously, the −100p main idea of the WT is to look at a −400p 22.5 23.0 23.5 24.0 24.5 signal at various windows and anaTime (s) lyze it with various resolutions. 300p Subsequently, the signal is fed through 100p an absolute value circuit, followed by a −100p −300p peak detector, to generate an adjustable 22.5 23.0 23.5 24.0 24.5 threshold level. It has been shown in [30] Time (s) that the two maximas with opposite signs 200p of the WT correspond to the QRS com100p 0.00 plex. The final signal-processing block is −100p −200p a comparator in order to detect the mod22.5 23.0 23.5 24.0 24.5 ulus maxima position. The time localizaTime (s) tion of the modulus maxima and the 80.0p 20.0p classification of characteristic points of −40.0p the cardiac signal are processed by digi−100p 22.5 23.0 23.5 24.0 24.5 tal logic circuitry. The wavelet system allows for easy Time (s) decomposition of the IECG signal (c) into components appearing at different scales (or resolutions) using short 500p 400p windows at high frequencies and long 300p windows at low frequencies. This fea200p 100p ture can be used to distinguish cardiac 0.00 24.1 24.5 22.5 signal points from severe noise and 23.3 22.9 23.7 interferences, as one can see in Figure Time (s) 14. Figure 14(b) shows a typical ven(d) tricular signal with 50-Hz interference (input signal), and Figure 14(c) gives Fig. 14. (a) Ventricular signal. (b) Ventricular signal with 50 Hz interference. (c) The the WT at various scales. We can see wavelet transform at five subsequent scales. (d) QRS complex modulus maxima in Figure 14(d) that both modulus detection for j=4. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

47

of cardiac signals using wavelets are being developed. Research on concepts such as the mathematical modeling of cardiac signals and pathologies and the design of WT-based algorithms for intelligent sensing and feature extraction are in progress. It remains to be demonstrated in clinical practice that these novel signal analysis methods will contribute to the further development and application of dynamical electrocardiography. It is clear, however, that implantable devices can document arrhythmias that typically escape from more conventional means of diagnosis. If they should be a method of last resort or if they should be applied early in the diagnostic chain deserves further clinical research including cost-effectiveness considerations. Conclusions

A brief overview of the history and development of circuit designs applied in pacemakers has been presented. The advances in IC designs have resulted in increasingly sophisticated pacing circuitry, providing, for instance, diagnostic analysis, adaptive rate response, and programmability. Also, based on future trends for pacemakers, some features and improvements for modern cardiac sensing systems have been described. Sandro A.P. Haddad received his B.Eng. degree from the University of Brasília, Brazil, in 2000, with honors. In February 2001, he joined the Electronics Research Laboratory, Delft University of Technology, the Netherlands, where he started research towards his Ph.D. degree. His project is part of BioSens (Biomedical Signal Processing Platform for Low-Power Real-Time Sensing of Cardiac Signals). His research interests include low-voltage, ultralow-power analog electronics and biomedical systems, and high-frequency analog integrated circuits for ultrawideband (UWB) communications. Richard P.M. Houben received a B.S. degree in electrical engineering from the Technische Hogeschool, Heerlen, The Netherlands, in 1984. From 1984–1989 he was engaged in industrial research and development of ultrasound imaging systems and the analysis of coronary angiograms. Since 1989, he has worked as a scientist at the Medtronic Bakken Research Center, Maastricht, the Netherlands. His current research involves processing of biomedical signals, especially focussing on ECG analysis and spatio-temporal analysis of atrial fibrillation. Wouter A. Serdijn received his ingenieurs (M.Sc.) and Ph.D. degrees from Delft University of Technology, the Netherlands, in 1989 and 1994, respectively. He is currently an associate professor at the Electronics Resesarch Laboratory of the same university. His research interests include low-voltage, ultralow-power, high-frequency, and dynamic-translinear analog integrated circuits (ICs) along with circuits for radio frequency (RF) and ultrawideband (UWB) wireless communications, hearing instruments, and pacemakers. 48 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

In these areas he authored and coauthored more than 150 publications and presentations. He teaches analog electronics, micropower analog IC design, and electronic design technicques. In 2001 and 2004, he received the EE Best Teacher Award. Address for Correspondence: Sandro A.P. Haddad, Electronics Research Laboratory, Faculty of Electrical Engineering, Mathematics and Computer Science, Delft University of Technology, Mekelweg 4, 2628 CD Delft, the Netherlands. E-mail: [email protected]. References [1] D.J. Woollons, “To beat or not to beat: The history and development of heart pacemakers,” Eng. Sci. Educ. J., vol. 4, no. 6, pp. 259–268, Dec. 1995. [2] L.A. Geddes, “Historical highlights in cardiac pacing,” IEEE Eng. Med. Biol. Mag., vol. 2, no. 2, pp. 12–18, Jun. 1990. [3] W. Greatbatch and C.F. Holmes, “History of implantable devices,” IEEE Eng. Med. Biol. Mag., vol. 10, no. 3, pp. 38–49, Sep. 1991. [4] R.S. Sanders and M.T. Lee, “Implantable pacemakers,” Proc. IEEE, vol. 84, no. 3, pp. 480–486, Mar. 1996. [5] H.G. Mond, “Recent advances in pacemaker lead technology,” Cardiac Electrophysiol. Rev., vol. 3, no. 1, pp. 5–9, 1999. [6] A.C. Guyton and J.E. Hall, Textbook of Medical Physiology, 9th ed. Philadelphia, PA: Saunders, Jan. 1996. [7] J.D. Bronzino, The Biomedical Engineering Handbook, 2nd ed. Boca Raton, FL: CRC, 2000, vol. 1. [8] R. Sutton and I. Bourgeois, The Foundations of Cardiac Pacing, Part I. Mt. Kisco, NY: Futura, 1991. [9] M. Schaldach and S. Furman, Advances in Pacemaker Technology. New York: Springer-Verlag, 1975. [10] A.S. Hyman, “Resuscitation of the stopped heart by intracardial therapy,” Arch. Intern. Med., vol. 50, p. 283–285, 1932. [11] Heart Rhythm Foundation, “Electricity and the heart: A historical perspective” [Online]. Available: http://www.hrsonline.org/ep-history [12] R. Elmqvist and A. Senning, An Implantable Pacemaker for the Heart, C.N. Smyth, Ed. London, UK: Tliffe & Sons, 1959, pp. 253–254. [13] W. Greatbatch, “Medical cardiac pacemaker,” U.S. Patent 3 057 356, Oct. 1962. [14] Integrated Publishing, Electrical engineering training series [Online]. Available: http://www.tpub.com/content/neets/14185/css/14185_125.htm [15] B.V. Berkovits, “Demand pacer,” U.S. Patent 3 528 428, Sept. 1970. [16] B.V. Berkovits, “Atrial and ventricular demand pacer,” U.S. Patent 3 595 242, July 1971. [17] J.W. Harthorne, “Pacemakers and store security devices,” Cardiol. Rev., vol. 9, no. 1, pp. 10–17, 2001. [18] C.C. de Cock, H.J. Spruijt, L.M. van Campen, A.W. Plu, and C.A. Visser, “Electromagnetic interference of an implantable loop recorder by commonly encountered electronic devices,” Pacing Clin. Electrophysiol., vol. 23, no. 10, Pt. 1, pp. 1516–1518, 2000. [19] T.W. Dawson, K. Caputa, M.A. Stuchly, and R. Kavet, “Pacemaker interference by 60-Hz contact currents,” IEEE Trans. Biomed. Eng., vol. 49, no. 8, pp. 878–886, 2002. [20] G. Lauck, A. von Smekal, S. Wolke, K.C. Seelos, W. Jung, M. Manz, and B. Luderitz, “Effects of nuclear magnetic resonance imaging on cardiac pacemakers,” Pacing Clin. Electrophysiol., vol. 18, no. 8, pp. 1549–1555, 1995. [21] S.A.P. Haddad, R. Houben, and W.A. Serdijn, “An ultra low-power dynamic translinear cardiac sense amplifier for pacemakers,” in Proc. IEEE Int. Symp. Circuits Systems, Bangkok, May 2003, vol. 5, pp. 37–40. [22] I. Daubechies, Ten Lectures on Wavelets. Philadelpha, PA: SIAM, 1992. [23] M. Jansen, Noise Reduction by Wavelet Thresholding. New York: Springer Verlag, 2001. [24] S.G. Mallat, A Wavelet Tour of Signal Processing, 2nd ed. San Diego, CA: Academic, 1999. [25] D. Donoho and I.M. Johnstone, “Adapting to unknown smoothness via wavelet shrinkage,” J. Amer. Statistical Assoc., vol. 90, no. 432, pp. 1200–1224, 1994. [26] C.D. Swerdlow, M.L. Brown, K. Lurie, J. Zhang, N.M. Wood, W.H. Olson, and J.M. Gillberg, “Discrimination of ventricular tachycardia from supraventricular tachycardia by a downloaded wavelet-transform morphology algorithm: A paradigm for development of implantable cardioverter defibrillator detection algorithms,” J. Cardiovasc. Electrophysiol., vol. 13, no. 5, pp. 432–441, 2002. [27] G. Strang and T. Nguyen, Wavelets and Filter Banks. Cambridge, MA: Wellesley-Cambridge, 1996. [28] S.A.P. Haddad, R. Houben, and W.A. Serdijn, “Analog wavelet transform employing dynamic translinear circuits for cardiac signal characterization,” in Proc. IEEE Int. Symp. Circuits Systems, Bangkok, May 2003, vol. 1, pp. 121–124. [29] S.A.P. Haddad, N. Verwaal, R. Houben, and W.A. Serdijn, “Optimized dynamic translinear implementation of the Gaussian wavelet transform,” in Proc. IEEE Int. Symp. Circuits Systems, Vancouver, Canada, 23–26 May 2004. [30] J.S. Sahambi, S.N. Tandon, and R.K.P. Bhatt, “Using wavelet transform for ECG characterization,” IEEE Eng. Med. Biol. Mag., vol 16, pp. 77–83, no. 1, Jan./Feb. 1997.

MAY/JUNE 2006

Locomotion Techniques for Robotic Colonoscopy

BY IRWAN KASSIM, LOUIS PHEE, WAN S. NG, FENG GONG, PAOLO DARIO, AND CHARLES A. MOSSE

ancer of the large intestine, also known as colon cancer, develops from the inner lining of the intestine walls and begins with benign polyps. These premalignant growths may eventually increase in size and become cancerous. As the cancer cells multiply, they form a tumor that bulges into the passage of the intestine and blocks movement of stools. The cancer cells can spread to other parts of the body to form new tumors. Low fiber intake, irregular toilet practices, and genetic factors affect the colon, resulting in irritable bowel syndrome, diverticulosis, colorectal cancer, rectal prolapse, hemorrhoids, and inflammatory bowel disease. Symptoms associated with colon cancer are blood in stools, changes in bowel habits, abdominal pain, weight loss, lumps in the abdomen, and bloody mucus discharge. Colon cancer can be treated successfully and even cured if diagnosed early. Unfortunately, many people feel uncomfortable discussing their bowel movements, so they often don’t seek medical attention until it is too late. Colon cancer can affect people at any age; however, the incidence rate rises sharply for those above the age of 50. Therefore, the emphasis is to educate people about colon cancer and the steps to reduce the risk of contracting the disease. One method is to maintain a high-fiber and low-fat diet and another is to be aware of its symptoms. It is encouraged that one should go for a full evaluation of the colon and rectum if any of the symptoms is experienced. This allows abnormalities to be detected early.

C A Literature Review

Conventional Colonoscope

Man has devised many instruments to help him inspect the rectum and colon. In the early 1900s, the first attempts were made with a rigid lighted telescope. With breakthroughs in fiber-optic technology, which allows light and images to be transmitted through curved structures, the first flexible colonoscope and sigmoidoscope were invented in 1963. This allowed an endoscopist to inspect the entire colon and to treat colonic diseases without the need for open surgery. In 1969, Olympus built the first commercial colonoscope [1]. Since then, colonoscopy has become a routine procedure in many hospitals all over the world. The conventional colonoscope consists of three main sections: umbilical cord, control unit, and insertion tube, Figure 1(a). The umbilical cord attaches the scope to the light source. The control unit houses the angulation knob and valves, which control the flipping of the distal end and air/water channel respectively. The insertion tube is the portion of the colonoscope that is IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

49

introduced into the patient’s colon via the anus. Running through all the three sections are the wires for a charge-coupled device (CCD) camera, an air/water channel, and fiberoptic bundles for light transmission. The insertion tube, however carries two extra items: the biopsy channel, which allows miniature surgical tools to pass through and control wires that are fixed to the angulation knob to enable flipping the distal end of the scope. At the distal end of the scope, all five items are arranged to sit into a headpiece of about 14 mm in diameter, as shown in Figure 1(b). Apart from just inspecting the colon walls during a colonoscopy, simple operations can be performed by passing miniature surgical tools via the biopsy channel of the scope. Colonoscopy Examination

Colonoscopy is one of the most technically demanding endoscopic examinations and is unpopular with patients. It is an art to coax an almost 2-m-long flexible tube around a tortuous colon while causing minimal discomfort and yet performing a thorough examination. Conventional colonoscopy involves manual insertion and maneuvering of the colonoscope, mainly the insertion tube, by the surgeon. A typical colonoscope is designed to be gripped in the left hand with the left thumb operating the angulation knob to control the flipping of the distal end and the left index finger operating the suction and air/water valves. The right hand is mainly used to advance the insertion tube. Most colonoscopists use similar endoscopic techniques. Air is pumped into the colon to distend it and aid insertion. The insertion pressure on the device must be gentle to avoid stretching the colonic wall, which can cause pain or perforation. The colonoscope is advanced by the pushing action of the endoscopist’s hand under direct vision. A variety of inand-out maneuvers are used to “accordion” the colon on the colonoscope, keeping the colonoscope as free of loops as possible. However, pushing is not the only action involved. Considerable skill is required to pull, wriggle, and shake the colonoscope. The patient’s abdomen may be pressed to minimize looping and discomfort. In a difficult colon, special maneuvers (like reducing the alpha loop in the sigmoid colon) are used to pass the sharply angulated sigmoid/descending colon junction. Torquing of the colonoscope is also required in such a scenario.

The detailed examination of the mucosa is performed both as the colonoscope is introduced and when it is slowly removed from the cecum. If the colonoscope is kept free of loops, the tip responds well and the examination is facilitated. This is especially true if a therapeutic procedure (such as polypectomy) is to be undertaken because large, redundant loops of the colonoscope can make control of the tip very difficult. Drawbacks of Conventional Colonoscopy

The invention of the colonoscope was a quantum leap towards noninvasive surgery. However, there is still room for further improvement as considerable drawbacks, mainly about the manipulation aspect of the procedure, are encountered: ➤ maneuvering difficulty due to manual insertion of the scope ➤ possibility of loop formation of the colonoscope shaft due to external insertion force ➤ demands a high level of expertise and a long training period ➤ the long duration of colonoscopy procedure (average about one half hour) ➤ continuous stress on the hands of the colonoscopist ➤ cumbersome and tedious procedure. Need for Automation

The drawbacks highlighted revolve around the fact that a human being is responsible for the technically demanding task of traversing the colonoscope. These problems may be solved by automating the locomotion aspect of colonoscopy, which has the following advantages: ➤ automatic locomotion of the colonoscope inside the colon ➤ reduction of trauma and discomfort to the patient ➤ ease of training with removal of the manipulative aspect of the procedure ➤ comparable time taken as in manual colonoscopy ➤ facilitatation of massive screening. With automation, the manipulating skills of the surgeon will no longer be the dependent factor. Instead, movement of the colonoscope will be controlled by computers, leading to a faster, more precise, and consistent motion. The endoscopist however, must be present at all times, to guide the machine to do its job. He is still the person who will decide every move the machine makes and to take over when there are uncertainties or in Umbilical case of an emergency. Cord

Air/Water Feeding Button Suction Button

Angulation Knob

Light Source Connector

Lens Water/Air Nozzle

Light Guide Diopter Adjustment Ring Guide

Eyepiece Insertion Tube (a)

Biopsy Channel

(b)

Fig. 1. The (a) conventional colonoscope and (b) its distal end view displaying the arrangement of the CCD camera, air/water channel, fiber optic bundle and biopsy channel. The control wire is fixed internally onto the headpiece.

50 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Robotic Colonoscopy

Locomotion is an essential part of robotic colonoscopy. The robot must be able to propel itself from the anus right up to the cecum without damaging the colon walls. The challenge is to design a robust locomotion technique that is able to advance through the stretchable, slippery, and mobile colon, which is always in its collapsed MAY/JUNE 2006

stage, in three-dimensional (3-D) orientation. In addition, it must have the ability to carry optical fibers, surgical tools, and other instruments required in a colonoscopy procedure. A steerable distal end should also be included to flip the robot’s tip, thus allowing the endoscopist to perform a more thorough examination and also position a surgical tool if surgery is required. The problems inherent to pushing a colonoscope through the colon have led to many proposals of providing propulsion at the tip so that the colonoscope can pull itself through the colon or move under natural contraction from the colon wall. On the other hand, there are many lessons to be learned from the animal world. Earthworms, starfish, caterpillars with suction pads on their feet, snakes, millipedes, quadrupeds and squids, or octopuses provide inspiration to imitate their locomotion style. Mechanical means such as wheels and pulleys were also proposed for propelling the endoscope. Earthworm Locomotion

showed that suction alone did not generate sufficient adhesive forces and slipping often occurred. However, it was noted that when a vacuum is introduced, the colon collapses to take the shape of any hard object within. A clamping concept was introduced to compliment the suction, as shown in Figure 2(a). Vacuum is introduced to cause the colon to collapse into the jaws of a mechanical clamp, after which the jaws close to grip onto the trapped tissue. A prototype with suction and clamping was built [Figure 2(b)]. Animal tests have shown that this prototype was capable of propelling itself more than 60 cm into the colon, reaching past the sigmoid colon [7]. Its greatest drawback was what the inventors termed the accordion effect. During the elongation or retraction phases of the inchworm’s locomotion, the colon wall may elongate or retract together with the inchworm; therefore no advancement will take place. A new prototype was designed to overcome this problem [Figure 2(c)]. Suction and a mechanical clamping concept was used; however, the clamps are arranged so that each pair of clamps can slide within each other. In doing so, they would always grasp onto the colon distal to what was previously grasped by the other pair of clamps. This locomotion concept is similar to how a monkey swings from tree to tree. Initial in vivo tests on animals have shown promising results.

The earliest and most common approach to propelling endoscopes is to simulate the way an earthworm moves. By alternately extending and distending sections of their body, an earthworm produces a peristaltic wave that drives it through the Snake Locomotion soil. The simplest earthworm locomotion technique, better Most snake species move by using their ventral scales, the known as inchworm motion, consists of two clampers at its scales on the undersides of their bodies, to pull themselves ends and one extensor at its midsection. The clamper is used to across rough surfaces. Even a paved road has enough rough adhere or clamp the device securely onto the colon wall, while spots for the ventral scales to gain a purchase and pull the the extensor brings about a positive displacement. The main snake along. Most species use a type of movement called serdifficulty of this technique is to be able to acquire sufficient pentine locomotion, in which the body assumes a position of a friction/grip to anchor the clamper onto the slippery colon wall. series of S-shaped horizontal loops, and each loop pushes In 1961, Drapier et al. [2] developed an automatic locomoagainst any surface resistance. tive device for a catheter using the inchworm locomotion techIn 1988, Ikuta et al. [8] developed an active endoscope that nique. He also introduced two methods of acquiring adhesion uses shape memory alloy (SMA) to guide a snakelike robot between the clamper and the colon wall; the suction and around obstacles. The SMA tendons were arranged around a expansion method. In both methods, the extensor is integrated spine so that each section can bend in three dimensions. The with an expansion spring to allow it to return to its original snake was operated manually via a joy stick controlling the length once the pressurized air to extend it is removed. two tip segments, and the tip bending instructions are then In 1979, Frazer [3] adopted a similar technique with the passed back along the line as the endoscope is then pushed expansion method of clamping and patented a design conforward so that subsequent sections follow their preceding segsisting of two radially expandable bladders, separated by an ment. The design comprises five segments, four of which are axially expandable bellows, with only the rear bladder flexible in the same direction on a plane and one segment, attached to the endoscope. Compressed air and vacuum was which is the tip, can bend orthogonally to this plane. The driproposed to respectively expand and contract both bladder ving mechanism of each segment consists of a stainless steel and bellows. Fukuda et al. [4] developed an in-pipe inspeccoil spring, which acts as the main skeleton at the center of a tion robot in 1989, which is capable of moving inside straight or bent pipelines of nuclear power stations. They believe that Elongation Clamping with proper miniaturizaMechanism Mechanism tion and human safety factor consideration, a similar robot design could be developed for automated colonoscopy. Dario et al. [5], [6] used suction to ascertain adhesion onto the colon (a) (b) (c) walls and developed a few prototypes to test its Fig. 2. Dario et al. [5]–[7], microrobotic endoscope: (a) a schematic diagram illustrating the workability. In vitro and mechanical clamping mechanism to grasp onto the tissue; (b) and (c) show the prototype with in vivo tests on animals both suction and mechanical clampers. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

51

joint, and a series of SMA coil springs arranged around it. In this model, one segment has one degree of freedom, so a pair of SMA actuators that are capable of antagonistic motion are arranged in symmetry with respect to the axis. It is this antagonistic activation of the SMA springs that brings about the required bending motion. Millipede Locomotion

Millipedes have many legs that move in waves, and the same principle can be applied to a colonoscope by incorporating legs so that it can be made to move back and forth, thus advancing the endoscope. Utsugi [9] proposed a walking system with three inflatable cuffs that form one section of the millipede, as shown in Figure 3(a). The middle, propellant cuff is the leg, which is pushed backward and forward by the cuffs on either side of it. The sequence of operation is that the propellant cuffs are inflated so that they press against the walls of the colon with enough force to prevent slipping. Next, the drive cuffs are inflated, thereby pushing the propellant cuffs backward so that the sheath, and hence the endoscope, moves forward. The return cuffs are now inflated so that they first lift the wall of the gut off the propellant cuffs and then push those cuffs back onto the deflated drive cuffs. The cycle is now complete, and one forward propelling step has been taken. In 1994, Allred [10] proposed an ingenious design that used washers as feet. In his design, as illustrated in Figure 3(b), the endoscope is surrounded by groups of five washers. All the

12

14

13

39

37 40 28 26 32 29

38

30 27

34

Drive Cuff

3111

33

41

Propellant Cuff

Return Cuff

(a) Fig. 3. (a) Utsugi [9] patent and (b) Allred [10] patent.

16

20 25 6d 5d 3d

6c 5c 10a 8 9 3c 10 7 12c 12d 10b 3b

11a

11

9a12b 12a 5b 6b 3a 1

5a

6a

Fig. 4. Treat et al. [11], four legged endoscope.

52 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

groups are connected together and move in unison, but within each group, every individual washer can be moved independently. Each of the five washers performs a cycle in which it moves slowly backward and then rapidly forward. If all five washers did this together, then the endoscope would simply rock back and forth. By forcing all of them to be out of phase such that, at any one time, four are moving slowly backward and only one is moving rapidly forward, the frictional force resisting skidding is independent of the speed of the skid. In this case, the forward propulsion from the four slow washers outweighs the reverse thrust from the one fast washer, and the endoscope slowly advances. Lizards and Ants Locomotion

Perhaps the most intriguing behavior about lizards is its ability to run right up vertical walls and across ceilings in an upside down manner. Their special toe scales, which stick to surfaces very effectively, gives them that advantage. Interestingly, ants display a similar behavior of climbing walls with ease. In 1997, Treat et al. [11] present a four-legged device with legs that can literally walk along the colon. The animal analogy is striking, and the creature has a single eye which sends a video image out through its tail to the endoscopist [Figure 4]. The legs of the robotic endoscope comprise four pneumatic linear actuators. When activated, self-propelled motive force is produced by pushing against the inner surfaces of the colonic wall. By activating the actuators in sequence, the robot can traverse by itself. Backward motion 112 can be achieved by pivWasher 112 oting the actuators to point in the opposite direction. However, due to the rigidity of the robot, it may not be flexible enough to negotiate the bends found in the colon. 114 Adopting a similar 110 concept, Phee et al. (b) [12], [13] embraced the idea further to develop the EndoCrawler. It is made up of five rigid segments joined together by a passive flexible rubber bellows. Four rubber bellows actuators are connected (90◦ apart) to each segment. A hollow central cavity, which runs through the whole length of the robot, houses the air/water tubing, optical fibers, CCD cables, and the biopsy channel. A steerable distal end was integrated to allow flipping of the scope during colon inspection. When air is introduced into the bellows actuators, it extends longitudinally until it touches the colonic wall at 45°, thus creating a resultant force that pushes the robot forward. When vacuum is introduced, the bellows will collapse. By activating the actuators at specific intervals, a variety of gait sequences are achieved. Different gait sequences can be achieved from the EndoCrawler; however, different gait sequences will perform better under different conditions. For example, for a gait sequence activated antagonistically, it performs well in a horizontal environment but fails if the path is vertical, against gravity. Therefore, a robust gait sequence should be ascertained in order to locomote inside the colon. MAY/JUNE 2006

Octopus Locomotion

An octopus escapes from its predators by squeezing water from its mantle and jetting away. The physical principle is that a mass is accelerated through an orifice and the resultant force produces a reaction that pushes the octopus in the opposite direction. It is the same principle that drives a rocket or a jet engine. Ginsburgh et al. [14] proposed that this principle could be used to propel a borescope. As shown in Figure 5(a), the tip of a borescope was attached with a tube (42 on the figure) that supplies pressurized liquid to the rear facing nozzle (40). As fluid is passed through the pipe, thrust in the opposite direction is generated as the fluid accelerates through the nozzle. As a result, the borescope advances. In 1998, Mosse et al. [15], [16] further developed the water jet propulsion method by constructing prototypes and testing them in models. Water is pumped to a headpiece at the tip of the scope where it is sprayed out of backward facing nozzles [Figure 5(b)]. As the water accelerates through the nozzles, the inertia force propels the endoscope tip forward. An in vivo experiment was conducted with a live pig using a water flow rate of 5 L/min and a pump pressure of about 20 Bar. The prototype was able to travel up to about 300 mm proximal to the anus. The wastewater flowed harmlessly out of the anus and could be easily collected. Despite pulling and pushing the endoscope along the bowel about a dozen times, the bowel was not damaged, although considerable redness and soreness could be seen. Locomotion Using Telescopic Technique

et al. [18]. For a forward motion, the head solenoid is activated, thus causing the permanent magnet to move toward it. Upon impact, a force is generated, which moves the actuator forward. After impact, the current direction of the head solenoid changes, therefore a repulsive force is induced onto the permanent magnet, causing it to move backward. The actuator moves forward continuously by repeating the sequence. Two prototypes of different sizes were built to verify the locomotion idea, and both prototypes move in accordance to the design. The larger size prototype advances faster than the smaller prototype due to its higher inertia. However, the design does encounter setback similar to that of the water jet principle; inadequate force produced to power the robotic colonoscope. The heat generated to magnetize the body should also be contained within a safe margin guideline of the health organization. Locomotion Using Natural Peristalsis

Modern endoscopes allow routine inspection of the upper and lower regions of the abdomen, yet the small bowel can still only be inspected with great difficulty. If a camera, light source, transmitter, and power supply could be made small enough to fit into a capsule that could be swallowed, then pain-free endoscopy would be possible. In 1997, Iddan et al. [19] patented an idea describing a swallowable capsule integrated with a minute camera system, light, and power supply. The camera is housed inside a capsule whose front portion is a transparent cone, and the image transmitted is of the mucosa sliding over this transparent cone as peristalsis pushes the capsule through the intestines. In 2000, Gong et al. [20] developed a dual capsule wireless endoscope prototype, which houses a miniature CCD camera, a processor, and a halogen light in one capsule and a microwave transmitter and batteries in another. The prototype was surgically inserted into the stomach of anaesthetized pigs. Using receiving aerial augmentation, good quality color images were received, showing detail of the stomach wall and contractions of the pylorus. The prototype

In 1976, Masuda [17] filed a patent proposing that a flexible fiberscope (item 36 in Figure 6) could be fed through a conduit by attaching it to the end of an everted tube, i.e., a tube whose end has been turned inwards and pulled back through itself. It can be seen that when the tube is filled with pressured liquid, it will unroll itself and pull the endoscope forward. Since the tube is rolling against the conduit wall, there is no sliding between it and the wall. From the dia42 gram, in the design there seems to be no annular 40 3 space to pass instruments 41 unless Item 35 is made A big enough. The process of unrolling looks simC 3 ple, but it must be remembered that in 3-D, B D the diameter is expanding as it rolls outward. If (a) (b) the everting tube is not to balloon out and stretch Fig. 5. (a) Ginsburgh et al. [14] patent, borescope propelled by a jet at its tip and (b) Mosse et al. the colon, then it must [15], [16] invention, a water jet propelled colonoscopy. not go on stretching once it has everted, which might be possible using a braided materi1 al. Such a material would have to be crumpled up before 34 33 3 everting, and unless it was extremely flexible, there might be 35 10 problems in unraveling it once the tube had been bent around the sigmoid colon and the splenic flexure. 17

36

Locomotion Using Impact

A robotic endoscope that locomotes by colliding a movable permanent magnet against two electrically magnetized stationary magnets on each side of a cylinder was proposed by Hyun IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 6. Masuda [17] patent, an everted tube is used to pull a fibrescope through a conduit.

MAY/JUNE 2006

53

was removed surgically after the experiment. This study demonstrates the feasibility of transmitting high-quality images through the abdominal wall with a small microwave and radio frequency transmitter. It was also the first to report successful wireless transmission of a moving color television image of the stomach in a living subject. An Israeli company, Given Imaging [21], developed the first commercial disposable capsulated pill named M2A. The single capsule design incorporates a light source, a miniature complementary metal oxide semiconductor (CMOS) color camera, a battery, an antenna, and a radio transmitter. Images captured by the video camera are transmitted by radio frequency to an array of sensors worn around the patient’s waist, where the signals are recorded digitally. No hospitalization is required when using M2A. The patient simply swallows the pill, puts the sensor around the waist (like a portable walkman), and proceeds with his daily affairs. After approximately eight hours, or after detecting that the capsule has been excreted, the patient removes the sensor and returns it to the clinic where the images are downloaded and the doctor examines the video to look for abnormalities. The entire process is painless and convenient for both patient and doctor. Still at its primarily stage, the Intelligent Microsystem Center [22] in Korea is developing a microcapsule named MiRO #1, as illustrated in Figure 7, which has the ability to

advance forward, backward, orientate, stop, and anchor itself onto the colon wall at will. Only 10 mm in diameter and 20 mm in length, it is equipped with a micromanipulator arm that is able to perform therapeutic procedures like taking tissue samples and administering an injection. The images are transmitted wirelessly, and it is also integrated with a position tracking device. Discussion

The primary objective of a robotic colonoscope is to propel itself from the rectum right up to the caecum without damaging the delicate colon walls, while carrying a camera, light source, biopsy channel, and air/water tubing. From reviewing other researchers’ work, clinical observations, and practical experience, the following characteristics are proposed to successfully develop a working robotic colonoscope. ➤ The body of the robot must be flexible enough to conform to the acute bends found in the colon. Any rigid distances must be kept to a minimum (max: 40 mm). ➤ The rigid diameter of the robot should not be greater than 29 mm, which is the smallest average internal diameter of the colon (rectal sigmoid). ➤ The robot’s body surface must be made of biocompatible material to eliminate any reaction between the robot and the colon tissue. ➤ A very experienced colonoscopist would take less than 5 min to reach the caecum, whereas a new hand would take 20–30 min. The robotic colonoscopy should be Micro Syringe Micro Optics placed in the “experienced” category, Temperature Sensor therefore, the authors feel that a benchpH Sensor mark of about 6–8 min for an automatic Chemical Sensor self-propelling robot to reach the caecum Stopping Mechanism would be optimal. ➤ The locomotion technique should not disMicro Tool Micro Pump turb the colon tissue and should keep to its RF Com. bare minimum to preserve the original integrity of the colon walls, which Mini Battery accounts for an accurate diagnosis and Signal Processor indirectly reduces discomfort experienced by the patient. Extension DC/DC Mechanism Figure 8 shows an overall picture of the litConverter erature reviewed in this article. It can be seen that most locomotion techniques depend on the colonic walls as a platform for the robot’s Fig. 7. Intelligent Microsystem Center [22], MiRO 1 Endoscopic Microcapsule. advancement. For instance, the animal

Locomotion Technique Mechnical and Other Method

Animal Locomotion Earthworm

Snake

Millipedes Lizard and Ants

Utsugi Drapier et al. Ikuta et al. Sturges et al. Allred Frazer Fukuda et al. Burdick et al. Vijayan et al. Dario et al.

Treat et al. Phee et al.

Octopus

Wheel and Pully Telescopic Impact

Ginsburgh et al. Mosse et al.

Takada Goh et al.

Natural Peristalsis

Masuda Hyun et al.

Iddan et al. Gong et al. Given Imaging Intelligent Microsystem Center

Fig. 8. An overall picture of the literature reviewed in this article.

54 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

locomotion method (earthworms, lizards, and ants) uses the basic principle of establishing a base to anchor part of its robot body onto the colon walls in order to push itself forward. The bases were achieved by either stretching the colon walls until they can no longer stretch further, sucking the colon walls (causing them to wrap the robotic element), physically clamping the colon wall with movable jaws, or digging into the colonic walls. At some portion of the colon, where the surrounding organs are relatively hard, e.g., the kidney, the amount of stretching reduces since these organ resist the colon movement, thus creating a resultant force to transverse the robot. The propulsion method mimicking a millipede, snake, and using mechanical means (wheels and pulleys) adopt the traction principle to advance the robot by ascertaining friction between the robotic element and the colon walls. However, due to the nature of the colon, which is highly elastic and slippery due to the presence of mucus, establishing an anchor area or traction is very difficult. Once the anchor area or traction is ineffective, mainly due to slippage or insufficient friction, no advancement will occur. This is normally what happened where the robot executed its manipulation sequence without achieving any locomotion. Perhaps the most promising work developed from these locomotion techniques is the inchworm method that uses suction and a mechanical clamp, integrated with a monkey style of swinging. Since it ensures a mechanical grip onto the colon walls and a positive advancement of the robot, slippage is very much reduced. Another approach to locomotion is where the locomotion technique is independent of the colon wall; for example, the octopus, telescopic, impact, and natural peristalsis principles. The octopus propulsion principle is a practical method of delivering propulsion, however, the question of force becomes an issue. If the required propulsion force is low, say 1 N, then this method is useful. However in the case of a colonoscopy, the force required is much larger. Increasing the force requires increasing the mass flow rate or the velocity of the water, which would compromise the safety and risk of the patient and are much harder to assess. In view of these difficulties, the jet propulsion idea has to be carefully reviewed. The telescopic method using the everting tube is a very simple design; it does not slide against the colon wall and keeps on increasing in length by introducing pressure inside the annular space. However, the inherent need for the tube to straighten when elongating indirectly implies straightening of the colon, which is not permitted in the human body. The impact principle has a similar downfall as the octopus principle, the driving force is inadequate to pull it through the colon. Locomotion through natural peristalsis using a wireless, swallowable capsulated endosopy is designed mainly for screening of the small intestine. In the large bowel, which is normally in its collapsed stage, the tissue would likely to wrap around the capsule, thus hindering the visualization, and would probably miss small polyps. If the capsule is designed to inflate, then the capsule can increase its size when entering the large intestine to detect abnormalities in the colon. With the implementation of a micromanipulator arm and the ability to control the capsule’s movement and orientation at will, this would be a huge advantage when evaluating the small intestine. Both diagnosis and therapeutic procedure can be performed simultaneously. The ability to move against the natural peristalsis of the gastrointestinal (GI) track would be ideal for colonoscopy as a larger version of the capsule, appropriate for IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

colon, can be developed that can be inserted through the anus and moved to the caecum under the guidance of the doctor. The primary goal of researchers is to develop a locomotive technique that can efficiently travel in the colon. Once a locomotion method proves to be working, the safety issue will be addressed, and it varies from one design to another. Consideration such as air pressure, stretching of the colon walls, clamping or sucking force, and undesired marks on the colon walls are some of the parameters that must be evaluated. Conclusions

This review article discusses the work done by researchers in the quest to automate the colonoscopy procedure. The results of some of the researchers seem very promising. In vitro and in vivo experimentation have been carried out to prove the possibilities of a robot crawling along a patient’s colon, treating polyps as they are encountered. The authors believe that in the future, say 10–15 years, conventional colonoscopy will be revolutionized, giving way to robotics to assist doctors in conoloscope manipulation and performing therapeutic procedures and leaving doctors to concentrate on the diagnostic aspect of the procedure, which would encourage mass screening as more patients can be evaluated per session. Irwan Kassim received his B.Eng. degree from Queensland University of Technology, Australia, in 1997 and his M.Eng. from Nanyang Technological University (NTU), Singapore, in 2003, both in mechanical engineering. Currently, he is a research fellow at National Neuroscience Institute (NNI), Singapore, in association with the newly setup Advance Integrated Medical Systems (AIMS) Laboratory embarking on a robot for skull base surgery project. His work on robotic colonscopy project was during his term as a research associate and master candidature in Computer Integrated Medical Intervention Laboratory (CIMIL), NTU. He received a Silver Medal in Tan Kah Kee Young Investors’ Award in 2001 for his entry “Robotic Colonoscopy.” His research interests include medical robotics, mechatronics, and biomedical engineering. Louis Phee received his B.Eng. and M.Eng. from Nanyang Technological University, Singapore, in 1996 and 1999, respectively. He later obtained his Ph.D. from Scuola Superiore Sant’Anna, Pisa Italy in 2002. He is currently an assistant professor at the Nanyang Technological University, Singapore. His research interests include medical robotics and mechatronics in medicine. He is also the principal investigator of a few government-funded research projects. Wan S. Ng is an associate professor at the School of Mechanical and Production Engineering, Nanyang Technological University, and a fellow of IMechE (U.K.). He is also the head and founder of the Computer Medical Intervention Laboratory (CIMIL). Under his leadership, CIMIL obtained a number of research MAY/JUNE 2006

55

grants from the Singapore government to conduct research ranging from image processing and computer visualization to robotics. The projects seek to help surgeons in both learning and execution, which can lead to improved clinical outcomes. More can be found at http://mrcas.mpe.ntu.edu.sg. CIMIL actively collaborates with several consultants in the local hospitals as well as overseas research institutions such as Imperial College of London, the United Kingdom and Rochester University of New York. His research interest is in medical robotics, computer assistance in surgery, and the safety of medical robots. Feng Gong received his B.Sc. in precision engineering from Harbin Institute of Technology, China, in 1983; his M.Sc in engineering from the University of Warwick, the United Kigndom, in 1994; and a Ph.D. in medical physics and bioenginnering from the University of London (UCL), the United Kingdom, in 1999. He spent nearly nine years in the Department of Medical Physics and Bioengineering, UCL, where he worked on several projects that aimed to design, develop, and test miniature instruments for flexible endoscopy and to develop novel techniques for wireless endoscopy. In May 2000, he joined the School of Mechanical and Production Engineering, Nanyang Technological University, Singapore, as an assistant professor. His major research areas include endoscopic suturing techniques, endoscopic miniature instrument development, and wireless video capsule technology. Paolo Dario received his Dr. Eng. Degree in Mechanical Engineering from the University of Pisa, Italy, in 1977. He is currently a Professor of Biomedical Robotics at the Scuola Superiore Sant’Anna in Pisa. He also teaches courses at the School of Engineering of the University of Pisa and at the Campus Biomedico University in Rome. He has been a visiting professor at Brown University, Providence, Rhode Island, at the Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland, and at Waseda University, Tokyo, Japan. He was the founder of the ARTS (Advanced Robotics Technologies and Systems) Laboratory and is currently the coordinator of the CRIM (Center for the Research in Microengineering) Laboratory of the Scuola Superiore Sant’Anna, where he supervises a team of about 70 researchers and Ph.D. students. He is also the director of the Polo Sant’Anna Valdera and a vice director of the Scuola Superiore Sant’Anna. His main research interests are in the fields of medical robotics, mechatronics and micro/nanoengineering, and specifically in sensors and actuators for the above applications. He is the coordinator of many national and European projects, the editor of two books on the subject of robotics, and the author of more than 200 scientific papers (75 on ISI journals). He is an editor-in-chief, associate editor, and editorial board member of many international journals. Charles A. Mosse studied politics, philosophy, and economics at Oxford University and applied mechanics at Cranfield Institute of Technology before taking up an appointment of 56 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Medical Physicist at University College, London, the United Kingdom. From 1980–1988, he worked on blood-gas and pH monitoring using electrochemical and optical sensors. He next worked on an optical system for measuring facial morphology and a technique using serial computed tomography (CT) scans to make titanium plates for cranioplasty. In 1992, he left to work for a dairy packaging company but returned in 1996 to undertake a Ph.D. titled “Devices to assist the insertion of colonoscopes,” which was completed in 1999. Since then he has divided his time between working on endoscopic devices and on light-delivery systems for photodynamic therapy (PDT) and for interstitial laser photocoagulation (ILP). Address for Correspondence: Wan S. Ng, School of Mechanical Engineering and Production Engineering, Computer Integrated Medical Intervention Laboratory, 50 Nanyang Avenue 639798 Singapore. Phone +65 6790 4411. Fax: +65 6790 1859. E-mail: [email protected].

References [1] J.M. Church, Endoscopy of the Colon, Rectum and Anus: New York: IgakuShoin Medical Publishers, 1995. [2] M. Drapier, V. Steenbrugghe, and B. Successeurs, “Perfectionnements aux cathéters médicaux,” France Patent 1,278,965, 1961. [3] R.E. Frazer, “Apparatus for endoscopic examination,” U.S. Patent 4,176,662, 1979. [4] T. Fukuda, H. Hosokai, and M. Uemura, “Rubber gas actuator driven by hydrogen storage alloy for in-pipe inspection mobile robot with flexible structure,” Proc. IEEE Int. Conf. Robotics Automation, Scottsdale, AZ, 1989, pp. 1847–1852. [5] P. Dario, M.C. Carroza, B. Lencioni, B. Magnani, D. Reynaerts, M.G. Trivella, and A. Pietrabissa, “A microrobot for colonoscopy,” in Proc. 7th Int. Symp. Micro Machine Human Science, IEEE, Nagoya, Japan, 1996, pp. 223–228. [6] L. Phee, D. Accoto, A. Menciassi, C. Stefanini, M.C. Carrozza, and P. Dario, “Analysis and development of locomotion devices for the gastrointestinal tract,” IEEE Trans. Biomed. Eng., vol. 49, no. 6, pp. 613–616, 2002. [7] A. Menciassi, A. Arena, L. Phee, D. Accoto, C. Stefanini, G. Pernorio, S. Gorini, M. Boccadoro, M.C. Carrozza, and P. Dario, “Locomotion issues and mechanisms for microrobots in the gastrointestinal tract,” in Proc. 32nd Int. Symp. Robotics, Seoul, Korea, 2001, pp. 428–432. [8] K. Ikuta, T. Masahiro, and S. Hirose, “Shape memory alloy servo actuator system with electric resistance feedback and application for active endoscope,” IEEE Int. Conf. Robotics Automation, Philadelphia, PA, 1988, pp. 427–430. [9] M. Utsugi, “Tubular medical instrument having a flexible sheath driven by a plurality of cuffs,” U.S. Patent 4,148,307, 1979. [10] J.B. Allred, “Self advancing endoscope,” U.S. Patent 5,345,925, 1994. [11] M.R. Treat and W.S. Trimmer, “Self propelled endoscope using pressure driven linear actuators,” U.S. Patent 5,595,565, 1997. [12] S.J. Phee, W.S. Ng, I.M. Chen, F. Seow-Choen, and B.L. Davis, “Locomotion and steering aspects in automation of colonoscopy,” IEEE Eng. Med. Biol. Mag., vol. 16, no. 6, pp. 85–96, Nov. 1997. [13] S.J. Phee, W.S. Ng, I.M. Chen, F. Seow-Choen, and B.L. Davies, “Visual control aspects in automation of colonoscopy,” IEEE Eng. Med. Biol. Mag., vol. 17, no. 3, pp. 81–88, Mar. 1998. [14] I. Ginsburgh, J.A. Carlson, G.L. Taylor, and H. Saghatchi, “Method and apparatus for fluid propelled borescopes,” U.S. Patent 4,735,501, 1988. [15] C.A. Mosse, T.N. Mills, and C.P. Swain, “A water jet propelled colonoscope,” 6th United European Gastroenterology Week (UEGW), vol. 41, Suppl. 3, p. E2, 1997. [16] C.A. Mosse, C.P. Swain, G.D. Bell, and T.N. Mills, “Water jet propelled colonoscopy—A new method of endoscope propulsion,” Gastrointest. Endosc., vol. 47, p. AB40, 1998. [17] S. Masuda, “Apparatus for feeding a flexible tube through a conduit,” UK Patent 1,534,441, 1978. [18] J.M. Hyun, J.L. Hvung, M.L. Young, J.P. Juang, K. Byungkyu, and H.K. Soo, “Magnetic impact actuator for robotic endoscope,” in Proc. 32nd Int. Symp. Robotics, Seoul, Korea, 2001, pp. 1834–1838. [19] G.J. Iddan and D. Sturlesi, “In-vivo video camera,” U.S. Patent 5,604,531, 1997. [20] F. Gong, C.P. Swain, and T.N. Mills, “Wireless endoscopy,” Gastrointestinal Endoscopy, vol. 51, no. 6, 2000. [21] Given Imaging, “Expanding the scope of GI—M2A,” [Online]. Available: http://www.givenimaging.com [22] Intelligent Microsystem Center, “Endoscopic microcapsule,” [Online]. Available: http://www.microsystem.re.kr

MAY/JUNE 2006

Noncontact Measurement of Breathing Function

BY RAMYA MURTHY AND IOANNIS PAVLIDIS

e have developed a novel method for noncontact measurement of breathing function. The method is based on the statistical modeling of dynamic thermal data captured through a highly sensitive infrared imaging system. The air that is breathed out has a higher temperature than the typical background of indoor environments (e.g., walls). Therefore, the particles of the expired air emit at a higher power than the background, a phenomenon that is captured as a distinct thermal signature in the infrared imagery. There is significant technical difficulty in computing this signature, however, because the phenomenona has a very low intensity and is of a transient nature. To address the problem, we use an advanced statistical algorithm based on multinormal data representation, the method of moments, and the Jeffreys divergence measure. In experimental tests, we were able to correctly compute the breathing waveforms in eight infrared video clips of three subjects at distances ranging 6–8 ft. The results were compared with ground-truth data collected concomitantly with a traditional contact sensor. Our experiments demonstrated the promise of this modality, which may find applications in the next generation of contact-free polygraphy and in sleep studies. Monitoring of breathing function has applications in polygraphy, sleep studies, sport training, early detection of sudden infant death syndrome in neonates, and patient monitoring. Various contact measurement methods have been developed for estimating the breathing rate of a subject. Moody et al. developed a contact modality in which numerous electrocardiogram (ECG) electrodes and sensors are attached to the subject [1]. The principle of operation is based on the fact that the heart rate is typically modulated by breathing, a phenomenon known as sinus arrhythmia [2]. Therefore, a signal corresponding to the heart function contains breath information, which is filtered out using band-pass filters. As an improvement over the ECG method, the BioMatt method [3] was developed in Finland by a group of researchers who were studying sleep disorders. BioMatt performs measurements of vital signs, such as breathing and cardiac activity without electrodes. Initially, BioMatt could not distinguish motion that was due to breathing versus cardiac activity or body movement. Later, Larson developed a signal processing technique to separate out the components of the BioMatt signal [4]. Photoplethysmography (PPG) is a variant method of the ECG, developed to measure blood volume changes in living tissues by absorption or scattering of

W A Novel Method Using Infrared Imaging and Advanced Statistics

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

57

near-infrared radiation. This modality consists of an infrared light-emitting diode (LED) and a photodiode that can be clamped to the ear lobes, thumbs, or toes. It is advantageous because it is portable, compact, and needs very little maintenance. The measurement of blood volume changes by PPG depends on stronger absorption of near-infrared light by blood when compared to other superficial tissues [5]. The amount of backscattered light corresponds to the variation of the blood volume. As in ECG, the breath waveform is separated from the cardiac signal through various methods that have been developed [6], [7]. However, using heart function as a basis for acquiring the breathing waveform is unreliable since sinus arrhythmia is not present in all individuals. Control of cardiac activity by breathing depends on the age and medications administered to subjects. Other contact modalities are capable of directly measuring the breathing signal. An example of such modality is the abdominal strain gauge transducer [8], which is strapped around the subject’s chest and measures the change in thoracic or abdominal circumference while breathing. Another example is a thermistor that measures nasal air temperature variation as an indication of breathing [9]. The disadvantage of all the aforementioned technologies is that they require close contact with the subject, which, in certain cases, may be quite uncomfortable and awkward (e.g., abdominal transducer). A contact-free but active technology called radar vital signs monitor (RVSM) [10] was developed in 1996 to monitor the performance of Olympic athletes. The RVSM detects breathing-induced movement of the chest based on the Doppler phenomenon. It measures breath at distances of up to 15 ft behind an 8-in hollow concrete or wooden wall. A radar flashlight [11] was built to make use of this capability in assisting law enforcement personnel to detect individuals hidden behind walls. In 2000, RVSM was used in noncontact polygraphy [12]. The disadvantage of this technique is that motion artifacts corrupt breath signals, and specialized frequency filters need to be used to separate them. In 2000, infrared imaging proved its potential in deception detection when thermal image analysis was used by Pavlidis et al. to detect facial patterns of stress at a distance [13]. A little later, Pavlidis et al. used infrared imaging to compute periorbital perfusion as a replacement of the corresponding polygraph channel that uses finger contact sensing [14], [15]. The proposed use of infrared imaging for computing breathing function may also replace the corresponding polygraph channel that uses

the abdominal transducer. Incremental replacement of contact channels with noncontact ones may prove very effective in the field of polygraphy, where it is essential that subjects feel as comfortable as possible during examination. Moreover, highly automated, noncontact monitoring of breathing function may have a significant impact on certain biomedical applications. For example, in sleep studies, this new methodology will enable monitoring of sleep apnea with minimal or no wiring of the subject, potentially at his/her home and not in the lab. This will not only improve the subject’s comfort but also facilitate much more sustained monitoring than is currently feasible. The use of infrared imaging for measuring breathing function is based on the fact that the exhaled air has a higher temperature than the typical background of indoor environments. This creates a discriminating thermal signature that can be captured through an infrared imaging sensor. The phenomenon is quasiperiodic and can be quantified using either statistics or calculus. From the statistical point of view, one can model the breathing cycles as multinormal distributions—one with cold temperatures corresponding to inhalation and one with hot temperatures corresponding to exhalation. From the Calculus point of view, one can model the quasiperiodicity of breathing through Fourier analysis. In this article, we describe a statistically based methodology for quantifying breathing rate with infrared imaging data. Alternative methodologies, like Fourier analysis, can be used but are not addressed in our present work. Our goal is to open a new line of research by demonstrating the feasibility of monitoring breathing function in a highly automated and noncontact fashion. In this article, we describe briefly the physiology of breathing. Then we refer to the visual tracking mechanism that enables consistent breathing measurements in the presence of subject motion. Our breath visualization scheme, which is of paramount importance during the training phase of our measurement algorithm, is described, and we explain in great detail the statistical algorithm that performs the breathing measurement on the infrared imaging data. Next, we outline our experimental design and results, and we finally conclude by discussing the strong and weak points of our methodology, its prospects, and our planned work for the future. Breathing Function

58 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Signal (mV)

Signal (mV)

Respiration in a man involves three well-defined stages [16]. The first stage is breathing. It comprises inspiration, taking oxygenated air into the lungs, and expiration, discharging air that is rich in carbon dioxide. The second stage involves the transport of the 2 Inspiration Inspiration 20 oxygen to the cells of the body using Expiration Expiration Pause the heart and the vascular system. 1 Pause 10 The third stage is called cellular res0 0 piration; in this stage, oxygen is used in the process of generating −10 −1 energy for physiological activities. In our study, we are interested in monitoring breathing using 8 9 10 19 20 21 22 23 24 25 26 3 4 5 6 7 infrared imaging. The breathing Time (s) Time (s) cycle consists of inspiration, expi(a) (b) ration, and postexpiratory pause. Fig. 1. Output from a piezo-respiratory belt transducer showing the three breathing During quiet breathing, inspiration phases during (a) quiet breathing and (b) after exercise. begins due to negative pressure MAY/JUNE 2006

created inside the chest cavity by the contraction of the diaphragm. Expiration is a passive process where the air flow occurs due to the elastic recoil property of the lungs. The postexpiratory pause is caused when there is equalization of the pressures inside the lungs and the atmosphere. Breathing cycle is defined as the time interval between the beginning of inspiration and the end of postexpiratory pause. During quiet breathing, the breathing rate may vary from 12–20 breaths/min and, after physical activity, 30–40 breaths/min in healthy individuals. Figure 1(a) and (b) shows typical duration of the three phases during quiet breathing and after physical activity, respectively. During quiet breathing, the duration of postexpiratory pause is comparable to that of inspiration and expiration. After a person undergoes physical exertion, the postexpiratory duration reduces considerably and, in some cases, this phase may even cease to exist. Tracking the Region of Interest

We define as the region of interest (ROI) R the region in the background, where there is possible presence of respiratory airflow. It is in this small image region that our statistical algorithm is applied. The ROI is characterized by its size, shape, and position. Over time, the size and shape of ROI remain the same, but its position changes to cope with the subject’s motion (tracking). We have experimented with different ROI sizes, and we will give more details about the optimal determination of this parameter later. In this section, we will address the issues of ROI shape and dynamic positioning. For simplicity, R was chosen to be a rectangular region. Typically, subjects are breathing through the nasal cavity, which results in a downward airflow profile. Breathing through the mouth is less prevalent and results in horizontal airflow profile. In our dataset, we observed downward airflow profiles [Figure 2(a)] in seven video clips and a horizontal profile [Figure 2(b)] in one video clip. Hence, we chose a rectangular region R arranged in a longitudinal fashion to closely match the prevalent downward profile of airflow. Our experiments have also shown that this shape still worked quite well on the video clip featuring the horizontal airflow profile (97.74% accuracy). The respective aims of the initial positioning and tracking algorithms are to provide the user with an approximate position of the ROI in the vicinity of the nasal-mandible region and follow it automatically throughout the breathing rate computation process. Technical details of the initial positioning and tracking algorithms can be found in [17]. The initial positioning algorithm locates the tip of the subject’s nose and places the ROI underneath it, in a position that is between the nostrils and the mouth (Figure 3). This placement works for the typical monitoring scenario of a subject imaged at a side view. For different monitoring scenarios, the above heuristic approach tends to place R incorrectly. But, the graphical user interface (GUI) gives options to move the ROI around the image so it can be placed within the airflow with just a mouse click. In such adjustments, the user is aided by the breath visualization tool. Hence, the initial positioning of ROI is semiautomatic; the algorithm gives the approximate position of R, and then the user may need to move it to a better position within the field of the airflow. By contrast, the tracking algorithm is completely automatic since it tracks R with respect to the tip of the nose in subsequent video frames without any feedback from the user. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Visualization of Breath

Visualization is important for, among other reasons, adjusting the initial ROI position and for training the statistical algorithm. Since the thermal signature of breath is not very strong, we have to apply image processing techniques to visually perceive breath in infrared video frames. Specifically, we apply the following operations on the video clip frames: 1) Otsu’s adaptive thresholding [18] 2) differential infrared thermography (DIT) [19] 3) image opening [20]. Otsu’s adaptive thresholding is applied to segment the skin region from the background. Then, in the background region, we apply DIT to generate a breath mask of all pixels whose temperature has increased beyond a preset threshold. This operation makes sure that the colormap is applied only to expiration frames in which DIT senses an increase in ROI pixel temperatures above the preset value. The result is a highly contrasting effect of no color during inspiration versus vivid color during expiration [see Figure 3(a) and (b)]. As a final step, an image open operation is applied on the output binary mask of DIT to improve breath visualization. The visualization scheme works well only for a limited period, as the temperature distributions of inspiration and expiration tend to drift over time for physiological and other reasons. DIT cannot handle this distribution drift well. If it did, it could have also been used to measure breathing rate in the place of the more sophisticated statistical algorithm (see the following section). However, the short time window of good visualization performance is adequate for adjusting the initial ROI position and training the statistical algorithm. Statistical Methodology

Integral to breathing rate computation using infrared imaging is the labeling of frames as expiratory and

R R

(a)

(b)

Fig. 2. Respiratory airflow profiles: (a) downward and (b) horizontal.

(a)

(a)

(b)

(b)

Fig. 3. Visualization of breath during (a) nonexpiration and (b) expiration. The ROI is anchored just under the subject’s nose tip.

MAY/JUNE 2006

59

Monitoring of breathing function has applications in polygraphy, sleep studies, sport training, and patient monitoring.

nonexpiratory. For this purpose, we have adopted a statistical methodology based on multinormal distributions, the method of moments, and the Jeffreys divergence measure [21]. We identify two phases in the statistical method: training and testing. We describe both phases in subsequent sections. First, we ascertain the normal nature of the temperature distributions in the ROI for the various breathing stages (Figure 4). Therefore, we can represent ROI distributions by their mean µ and variance σ 2 only. Our method combines inspiration and postexpiratory pause phases, since the thermal signatures of these two are almost identical. We designate the combined stage as nonexpiratory phase. Training Phase

The algorithm runs through a training phase to generate estimates of the expiration and nonexpiration distributions from the first few video frames. These estimates are then used to label pixels as expiratory or nonexpiratory in the initial video frame of the testing phase. We use a variant of the K-means clustering algorithm [22] to generate training data. Our objective is to form K = 2 representative distributions through an iterative process; a hot one for expiration and a cold one for nonexpiration. Initially, we specify as the expiration distribution De,0 the one with the hottest mean temperature µe,0 in the training set; we specify as the nonexpiration distribution Di,0 the one with the coldest mean temperature µi,0 in the training set: 9

Expiration Pause Inspiration

8

Number of Pixels

7

µe,0 = max {µj},

(1)

µi,0 = min {µj},

(2)

1≤ j≤M 1≤ j≤M

where N(µj, σj2 ), 1 ≤ j ≤ M is the set of temperature distributions for ROI R corresponding to the first M = 100 training frames of infrared video. We sort the distributions in ascending order with respect to their means in order to facilitate the iterative process. On every step j, 1 ≤ j ≤ M, we find the statistical distance of distribution Dj ∼ N(µj, σj2 ) from De, j−1 and Di, j−1 . For this purpose, we use the Jeffreys divergence measure as follows:


2 σe, j−1 σj − σj σe, j−1   1 1 1 (µe, j−1 − µj)2 , (3) + + 2 2 σj2 σe, j−1
 σj 2 1 σi, j−1 J(Di, j−1 Dj) = − 2 σj σi, j−1   1 1 1 (µi, j−1 − µj)2 . (4) + + 2 2 σj2 σi, j−1

J(De, j−1 Dj) =

1 2

The Jeffreys divergence measure is a symmetric form of the Kullback-Leibler divergence measure. It is a function of the means and standard deviations of the two distributions being compared. Hence, it is an appropriate distance measure for bivariate distributions. We choose the winner distribution Dw, j−1 (w = e or i) at step j as the one whose Jeffreys distance from the training population Dj is the smallest. The mean and variance of the winning distribution are then updated at each step as follows:

6

µw, j−1 + µj , 2 2 2 σw, j−1 + σj 2 . σw, j= 2

µw, j =

5 4

(5) (6)

3 2 1 24.5

24.55

24.6 24.65 24.7 24.75 Pixel Temperature (°C)

24.8

24.85

Fig. 4. Experimental temperature distributions for expiration, postexpiratory pause, and inspiration. The normal nature of the distributions and overlapping between postexpiratory pause and inspiration are evident.

60 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

The loser distribution retains its previous values. We iterate this process for all the training populations except those that were marked as initial clusters. At the end of the process, we obtain the estimates of the two distributions corresponding to the expiration and nonexpiration phases of the breathing cycle. Figure 5 depicts our K-means training method. Testing Phase

During the testing phase, we represent at frame t each pixel xt in region R as a mixture of two distributions: MAY/JUNE 2006

 

2 + πi,t N µi,t , σi,t2 , f(xt ) ∼ πe,t N µe,t , σe,t

(7)

We compute the Jeffreys divergence measures between the incoming distribution g(θxt ) and the existing nonexpiration fi(xt−1 ) and expiration fe (xt−1 ) distributions, respectively. Specifically,

2 ) is the normal expiration distribwhere fe (xt ) ∼ N(µe,t , σe,t ution, fi(xt ) ∼ N(µi,t , σi,t2 ) is the normal nonexpiration distribution, and πe,t and πi,t are their respective weights in the mixture satisfying the criterion

J( fi(xt−1 ), g(θxt )) =

πe,t + πi,t = 1. In the beginning of the testing phase (t = 0), the distributions for nonexpiration and expiration are equiprobable with πe,t = πi,t = 0.5 and are parameterized by the respective means and variances that we computed during the training phase. Therefore, every pixel in region R is represented as having the following starting distribution:

J( fe (xt−1 ), g(θxt )) =

 σg,t 2 σi,t−1 − σg,t σi,t−1   1 1 1 (µi,t−1 − µg,t )2 , + + 2 2 2 σg,t σi,t−1 1 2





2

(9)

σg,t σe,t−1 − σg,t σe,t−1   1 1 1 (µe,t−1 − µg,t )2 . + + 2 2 2 σg,t σe,t−1 1 2

(10) 

2 2 + 0.5 N µe,0 , σe,0 . f(x0 ) ∼ 0.5 N µi,0 , σi,0

(8)

We consider that the incoming distribution is closer to the existing distribution that features the minimum Jeffreys divergence measure. We call this the winning distribution fw (xt−1 ) and the other the losing distribution fl(xt−1 ). Based on this information, we update the parameters of the mixture following the method of moments. Specifically, we update the weights for both distributions and the mean and variance of the winning distribution only. The mean and variance of the losing distribution remain the same.

At time t > 0 and for pixel xt , we compare the incoming temperature value from the sensor with the estimated distribution from the previous frame at time t–1. For this comparison to be effective, we consider that the incoming temperature θxt can be associated to a normal distribution 2 ) , where µg,t = θx,t and σg,t = NEDT. g(θxt ) ∼ N(µg,t , σg,t For the camera model that we use, NEDT = 0.01 ◦ C.

0

I T E R A T I O N ... i,e ...

1

ROI Distribution Initialize Centroid (Training Samples)

100 Compare

Compare

D100

D1

D1 Di,0 (µmin)

J(De,0,D1)

J(De,99,D100)

J(Di,0,D1)

J(Di,99,D100)

S

Di Di,0

De,0

K

Update the Winner

I

Di,1

P

De,99

Di,99

Update the Winner Di,100

De,0 (µmax)

De De,1

D100

De,100

Training Data

Fig. 5. Iterations in the K-means training data acquisition method. The resulting expiration and nonexpiration distributions at the end of the M=100th iteration are used to jump-start the testing phase.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

61

The weights of the winning and losing distributions are updated as follows:

Pixel Distribution Training Distribution

πw,t = (1 − ρ)πw,t−1 + ρ, πl,t = (1 − ρ)πl,t−1 .

(11) (12)

The mean and variance of the winning distribution are updated as follows:

Before Updating After Updating

µw,t =(1 − ρ)µw,t−1 + ρµg,t , 2 2 2 =(1 − ρ)σw,t−1 + ρσg,t σw,t

Fig. 6. Comparison of a training distribution before and after the application of the method of moments.

(13)

+ ρ(1 − ρ)(µg,t − µw,t−1 )2 .

(14)

The parameter ρ is a learning parameter that is computed from the following formula [23]: ROI Temperature Values Repeated for Every Pixel in the ROI Jeffreys Divergence Measure Computation J(fi,t–1,gt)

(µe,t–1,σ2e,t–1,πe,t–1)

Training Data

(µw,t,σ2w,t,πw,t,πl,t)

πe,t

πi,t

Decide on the Pixel Label and Increment Ci or Ce Accordingly Ci

Ce

Decide on the Frame Label

Fig. 7. Overview of the statistical methodology for labeling infrared video frames as expiration or nonexpiration.

Initial Run of Similar Labels

Cycle 1


1 2

(µg,t − µw,t−1 ) σw,t−1

2 .

(15)

(µi,t–1,σ2i,t–1,πi,t–1)

J(fe,t–1,gt)

Update Training Data Using Method of Moments

− 12 ρ=e

Figure 6 is a visualization example of how the incoming distribution may affect the existing distribution. The updated data acts as the new estimate for the corresponding pixel in the next incoming frame. The pixel xt is given the label of the distribution with the highest updated weight. A count is kept of the number of expiration Ce and nonexpiration Ci pixels in region R at time t. Once all the pixels in region R are processed, the frame gets the label of the most frequently occurring pixel label. For example, if Ci > Ce , the frame is labeled as nonexpiration; otherwise, the frame is labeled as expiration. Figure 7 shows the flow of control and data through the statistical algorithm. The breathing rate computation algorithm keeps track of the frame labels and continuously updates the time of the current breathing cycle Tc by using the current timestamp Tn and the previous timestamp Tn−1 . The

Cycle 2

Skip

Subcycle 1

Subcycle 1

Subcycle 2

Subcycle 2

Fig. 8. An example labeling of the video frame line and computation of the breathing rate.

62 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 9. Phoenix midwave infrared camera equipped with a 50-mm lens.

MAY/JUNE 2006

initial run of similar frame labels (Figure 8) is skipped since the testing phase might have started in the middle of a cycle. The algorithm keeps track of the two subcycles in the breathing cycle, and once it detects the beginning of the next cycle, the breathing rate of the current cycle is computed in cycles/min using the formula Rate =

60 cycles/min. Tc

along the timeline for one of the subjects in our dataset. From the plot, we observe that the ROI temperature increases by around 0.1 ◦ C during expiration. Figure 13 shows the plot of ROI variance along the timeline for one of the subjects in our dataset. From the plot, we observe that the ROI variance during expiration is quadrupled. This is because within the ROI, there are clusters of hot air molecules interspersed with cold

(16) Infrared Camera System

Experimental Setup

We used a cooled midwave infrared Phoenix camera with a spectral range of 3.0 − 5.0 µm (Indigo Systems, Goleta, California) equipped with a 50-mm lens (Figure 9). The focal plane array of the camera is FPA = 640 × 512 pixels in size and has thermal sensitivity NEDT = 0.01 ◦ C. An external black body (Santa Barbara Infrared, Santa Barbara, California) was used to calibrate the camera. Infrared video frames were acquired at a rate of 31 frames/s. We captured the profile view of the subject’s face and respiratory airflow from a distance of 6–8 ft (Figure 10). A piezo-strap transducer [Figure 11(a)] wrapped around the subject’s diaphragm measured the thoracic circumference during expiration and nonexpiratory phase. The transducer sent its signal to a PowerLab data acquisition system [ADI Instruments, Australia—Figure 11(b)]. This was the gold standard that we used for benchmarking the infrared imaging measurements.

ADI Piezo - Respiratory Belt Transducer

ADI PowerLab Data Acquisition System

Fig. 10. Experimental setup wiith the profile view of the subject’s face captured by a midwave infrared Phoenix camera and ground-ruth data recorded concomitantly using a piezo-strap transducer and a PowerLab data acquisition system.

Experimental Results

In this section, we investigate experimentally different aspects of the breathing function and our method’s parameters. We also describe the performance of our noncontact methodology against ground-truth measurements taken by the PowerLab/4SP ADI instrument. When air is breathed in, it gets warmed up during its passage into the respiratory system and during its brief stay in the lungs. Figure 12 shows the plot of mean ROI temperature

(a)

(b)

Fig. 11. (a) The piezo-strap transducer and (b) the PowerLab data acquisition system.

12

25.62

11

25.61

10 ROI Variance (°C2)

ROI Mean (°C)

× 10−4 25.63

25.6 25.59 25.58 25.57 25.56 25.55 100

9 8 7 6 5 4

200

300 400 500 Time (Frames)

600

700

Fig. 12. A plot of the mean pixel temperature in the ROI along the timeline.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

3 100

200

500 300 400 Time (Frames)

600

700

Fig. 13. A plot of the variance of pixel temperature in the ROI along the timeline.

MAY/JUNE 2006

63

Highly automated, noncontact monitoring of breathing function may have a significant impact on certain biomedical applications.

air resident in the room. The hot air molecules are the ones recently expired through the nostrils. Our experimental protocol called for measurements during the following phases: ➤ breathing while the subject rests in a chair ➤ breathing after the subject has stepped on and off a foot step 30 times in 30 s ➤ breathing after the subject rested for 10 min so that his physiology reverts back to baseline

➤ breathing after the subject has stepped on and off a foot

step 60 times in 60 s. We measured the performance of our method on eight thermal clips captured from three subjects. The thermal clips were 1,000 frames long. The first 100 frames of each clip were used for training and the remaining 900 frames for testing. We were able to perform the infrared imaging measurements for all four phases of the experimental protocol only for Subject 2. For the other two subjects, we had to discard some of the measurements because of technical problems in synchronizing the camTable 1. The accuracy of breathing rate measurements for three different era with the ADI vital signs monitor. sizes of ROI. The accuracy was determined by ground-truthing the results Since the performance of the method clearfrom our algorithm against concomitant measurements recorded with ly depends on the size of the ROI where the the ADI vital signs monitor. statistical computation is taking place, we have experimented with three different ROI Subject Video Clip Small ROI Medium ROI Large ROI sizes: small (7 × 3 pixels), medium (21 × 9 pixels), and large (63 × 27 pixels). From the 1 1 98.19 94.32 57.71 experimental results in Table 1, we observe 2 92.59 96.14 83.03 that the medium-size ROI outperforms the 3 97.50 97.74 97.31 other two sizes. The interesting fact is that there is a clear breakdown in performance 2 1 94.76 96.70 67.34 when the ROI size is getting large. In such a 2 97.91 93.36 78.26 case, a significant number of the ROI pixels are background and not expiratory pixels. As 3 99.23 99.31 56.65 a result, they bias the ROI labeling towards 4 98.06 99.05 97.95 nonexpiration, and the accuracy drops. The 3 1 87.71 94.82 67.62 absolute ROI sizes are of course dependent on the optics. In our specific experimental Average Accuracy (%) 95.74 96.43 75.73 scenario, we recorded from a distance of 6–8

35 Computed Rate Average Computed Rate Average Ground Truth Rate

60

Breath Rate (Cycles/min)

Breath Rate (Cycles/min)

30 25 20 15 10 5 0

50 40 30 20

0 1

2

3 4 5 6 7 Breathing Cycle Number

8

9

Fig. 14. Breathing rate measurements during the initial resting phase for Subject 2.

64 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Computed Rate Average Computed Rate Average Ground Truth Rate

10

0

2

4

6 8 10 12 14 Breathing Cycle Number

16

18

Fig. 15. Breathing rate measurements after 30 s of exercise for Subject 2.

MAY/JUNE 2006

Infrared imaging for measuring breathing function is based on the fact that exhaled air has a higher temperature than the typical background of indoor environments.

ft with a 50-mm lens, and the ROI sizes we cite are commensurate to this optical arrangement. Figures 14–17 show the variation of breathing rate computed through the infrared imaging method for the four phases of the experimental protocol for Subject 2. They also show the comparison between the mean breathing rate computed through the infrared imaging method and the mean ground-truth rate measured by the ADI vital signs monitor. From the plots, we observe that the breathing rate increases from 12–20 cycles/min during rest to 30–40 cycles/min after the brief exercise. The accuracy of the infrared imaging method appears to be consistent during rest as well as during active periods. The ground truth signal has output proportional to the expansion (positive-level signal) and relaxation (zero-level signal) of the breathing monitor belt during inspiration and expiration, respectively. The signal computed from the infrared imaging method has output labeled either as nonexpiration or expiration. To make the comparison between ground-truth data and algorithmic results easier, we have digitized both signals by assigning a zero-level signal to nonexpiration and a positive-level signal to expiration. In addition, we have assigned a negative signal level to frames used for acquiring training data. In Figure 18, we can observe that the cycles detected by the infrared imaging method are slightly out of phase with the ground-truth cycles. This accounts for the small discrepancy that exists between the measurements of the infrared imaging method (middle ROI) and the ground-truth instrument.

Three primary factors account for this phase shift: ➤ imperfect (manual) synchronization of the beginning of the

two recordings (infrared video and monitor belt) ➤ mismatch of recording frequencies (our infrared camera

records at 31 frames/s, while the monitor belt samples at 100 times/s) ➤ the monitor belt records ground-truth data at the diaphragm level, while our infrared imaging method classifies airflow at the nasal-mandible level. The first factor can be addressed by developing a hardware trigger. The second factor can be addressed by downsampling the ground-truth signal. The amount of phase shift due to the third factor can be determined and taken into account by performing further research in this direction. Conclusions

Breathing function is one of the major indicators of an individual’s health. It can be used to predict various life threatening disorders like sudden infant death syndrome and heart attacks. It is also used in sleep studies to detect sleep apnea. Finally, it is one of the psychophysiological channels in polygraph examinations. Various modalities have been developed to measure breathing rate. Almost all the legacy methods require contact; therefore, they compromise the subject’s comfort and mobility. Moreover, measurements by these methods are corrupted either by movement artifacts or by their dependence on other physiological variables, like heart rate. We have proposed a method that is based on infrared imaging and statistical

45

40 Computed Rate Average Computed Rate Average Ground Truth Rate

40 Breath Rate (Cycles/min)

Breath Rate (Cycles/min)

35 30 25 20 15 10

35 30 25 20 15 10

Computed Rate Average Computed Rate Average Ground Truth Rate

5

5

0

0 1

2

3

4 5 6 7 8 9 10 Breathing Cycle Number

11

Fig. 16. Breathing rate measurements after 10 min rest for Subject 2.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0

2

4

6 8 10 12 14 Breathing Cycle Number

16

18

Fig. 17. Breathing rate measurements after 60 s of exercise for Subject 2.

MAY/JUNE 2006

65

computation to passively measure breathing rate at a distance. This method achieved an accuracy of 96.43% on a small set of subjects during rest and after brief exercise. It has the potential to provide a unique capability for sustained monitoring of chronic or acute breathing problems and in sleep studies by overcoming the deficiencies of the existing measurement modalities. It also opens the way for the next generation contact-free polygraphy that will not affect the subject’s psychophysiology. Future Work

Since our method is contact free, it has significant advantages over contact modalities like ECG, PPG, nasal temperature probe, and breathing monitor belt in terms of comfort. With the use of a simple tracking algorithm, our method has overcome the drawback of active noncontact modalities, like the radar vital signs monitor, whose output gets corrupted by motion artifacts. But, our tracking algorithm cannot deal with situations wherein the ROI fails to remain in the field of respiratory airflow, which occur when the subject rotates his/her head towards or away from the camera (Figure 19) or the source of airflow (either the nose or the mouth) changes.

The first problem may be addressed by developing an advanced nasal-mandible tracking algorithm (Figure 20) along with further research on detecting the respiratory airflow signal in frontal views. The second problem can be addressed by using two ROIs, one each for nasal and mandible airflow (Figure 21). In our algorithm, we have made use of the Gaussian nature of the thermal signature of breath to develop a statistical algorithm that classifies the frames as expiration or nonexpiration. An alternative approach would be to consider the quasiperiodic nature of the thermal signal (Figures 12 and 13), which renders itself naturally to fast Fourier transforms. Although our system is meant to be used in climate-controlled environments like modern clinics and homes, an intriguing question is how it will perform in more challenging environmental conditions. In future studies, we will study carefully the effect of rapid temperature changes in the environment on the performance of our method. Theoretically, the adaptive statistical mechanism of the method is expected to cope well in most cases. It would be valuable, however, to establish experimentally the operational envelope of our system.

10 Ground Truth Data Digitized Ground Truth Data Computed Data

8

Signal (mV)

6

Possible ROI Position with Advanced Tracking Algorithm

4 2 0 −2 Expiration Phase Training Phase

−4 −6

0

Nonexpiration Phase

Fig. 20. Likely position of ROI if advanced tracking algorithms were employed.

100 200 300 400 500 600 700 800 900 1,000 Time (Frames)

Fig. 18. This plot shows the phase shift between the breathing signal (Subject 2, Clip 4) computed from the infrared imaging method and the correponding ground-truth signal. ROI 1 ROI 2

ROI Position with Our Tracking Algorithm

Fig. 19. ROI wrongly positioned when the subject turns his head towards the camera due to the loss of the reference point.

66 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Fig. 21. The problem of a change in source of respiratory airflow can be solved by using an ROI each at the nose and the mouth.

MAY/JUNE 2006

Future studies will be conducted with larger sample sizes and subjects having undergone various degrees of physical exertion. Such studies will establish how well the method scales up to the general population. Although, the answer to this question is pending, the current work clearly establishes for the first time the feasibility of monitoring breathing function in a contact-free manner and with totally passive means.

detection technology. He is a Fulbright Fellow, a Senior Member of the IEEE, and a member of the ACM. He also serves as associate editor for the journal Pattern Analysis and Applications (Springer), and he has chaired numerous major IEEE conferences. Address for Correspondence: Ramya Murthy, 3030 Hidden Mist Court, Pearland, TX 77584 USA. Phone: +1 713 436 9157. E-mail: [email protected].

Acknowledgments

This research was supported by the National Science Foundation (grant IIS-0414754), by DARPA (NSF grant N00014-03-1-0622), and by the University of Houston start-up funds of Prof. I. Pavlidis. The views expressed in this article do not necessarily reflect the views of the funding agencies. We are thankful to Dr. Ephrain Glinert at NSF and Dr. Ralph Chatham at DARPA for their support. We would also like to thank Dr. Arcangelo Merla for his valuable help in the human experiments and Dr. Panagiotis Tsiamyrtzis for many valuable discussions on advanced statistics issues. Ramya Murthy received an M.S. degree in computer science from the University of Houston and a B.E. degree in computer science and engineering from Bangalore University, India. She worked in the field of infrared imaging as a research assistant in the Computational Physiology Lab at the University of Houston, Texas. Before pursuing her M.S. degree, she worked as a software engineer at Robert Bosch India Limited in the field of embedded mobile communications. She is a scientific programmer in the Laboratories for Biocomputing and Imaging at the University of Texas Medical School in Houston, where she performs algorithmic research and develops software tools for two- and three-dimensional modeling and visualization of biomolecular structures. Ioannis Pavlidis holds M.S. and Ph.D. degrees in computer science from the University of Minnesota, an M.S. degree in robotics from the Imperial College of the University of London, and a B.S. degree in electrical engineering from Democritus University in Greece. He joined the Computer Science Department at the University of Houston, Texas, in September 2002. His current research interests are in computational medicine, where he is charting new territory. He has developed a series of methods to compute vital signs of subjects in an automated, contact-free, and passive manner. This new technology has found widespread applications in computational psychology and is expected to find additional applications in preventive medicine. The quantification of stress in particular, through the computation of periorbital blood perfusion, is his most well-known piece of research. It is this research that established him as one of the founders of modern lie

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

References [1] G.B. Moody, R.G. Mark, M.A. Bump, J.S. Weinstein, A.D. Berman, J.E. Mietus, and A.L. Goldberger, “Clinical validation of the ECG-derived respiration (EDR) technique,” Comput. Cardiol., vol. 13, pp. 507–510, 1986. [2] T. Kim and M.C.K. Khoo, “Estimation of cardio respiratory transfer under spontaneous breathing conditions: A theoretical study,” Amer. J. Physiol., vol. 273, no. 2, pp. H1012–H1023, Aug. 1997. [3] M. Partinen, J. Alihanka, and J. Hasan, “Detection of sleep apneas by the static charge-sensitive bed,” in Proc. 6th European Congress Sleep Research, Zurich, Mar. 1982, pp. 312. [4] B.H. Larson, “Signal processing techniques for non-invasive monitoring of respiration and heart rate,” M.S. thesis, Dept. Electrical Sci., Univ. Houston, May 1987. [5] D. Barschdorff and W. Zhang, “Respiratory rhythm detection with plethysmographic methods,” in Proc 16th Annual Int. Conf. IEEE/EMBS, Baltimore, MD, Nov. 1994, pp. 912–913,. [6] L.M. Vicente, A.B. Barreto, and A.M. Taberner, “DSP removal of respiratory trend in photoplethysmographic blood volume pulse measurements,” in Proc. IEEE Southeastcon ‘96–Bringing Together Education, Sci, and Technol., pp. 96–98, 98a, [7] A. Johansson, L. Nilsson, S. Kalman, and P.A. Oberg, “Respiratory monitoring using photoplethysmography—Evaluation in the postoperative care unit,” in Proc. 20th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., 1998, vol. 20, no. 6, pp. 3226. [8] K. Nepal, E. Biegeleisen, and T. Ning, “Apnea detection and respiration rate estimation through parametric modeling,” in Proc. 28th IEEE Annu. Northeast Bioengineering Conf., Philadelphia, PA, 20–21 Apr. 2002, pp. 277–278. [9] K. Storck, M. Karlsson, P. Ask, and D. Loyd, “Heat transfer evaluation of the nasal thermistor technique,” IEEE Trans. Biomed. Eng., vol. 43, no. 12, pp. 1187–1191, Dec. 1996. [10] E.F. Greneker, “Radar sensing of heartbeat and respiration at a distance with applications of the technology,” RADAR, vol. 97, no. 449, pp. 150–154, Oct. 1997. [11] E.F. Greneker and J.L. Geisheimer, “The RADAR flashlight three years later: An update on developmental progress,” in Proc. IEEE 34th Annu. Int. Carnahan Conf. Security Technol., 2000, pp. 170–173 [12] J. Geisheimer and E.F. Greneker III, “A non-contact lie detector using radar vital signs monitor (RVSM) technology,” in Proc. IEEE 34th Annu. 2000 Int. Carnahan Conf. Security Technology, pp. 257–259. [13] I. Pavlidis, J. Levine, and P. Baukol, “Thermal imaging for anxiety detection,” in Proc. 2000 IEEE Workshop Computer Vision Beyond Visible Spectrum: Methods and Applications, Hilton Head Island, SC, pp. 104–109. [14] I. Pavlidis, N.L. Eberhardt, and J. Levine, “Human behavior: Seeing through the face of deception,” Nature, vol. 415, no. 6867, Jan. 3, 2002. [15] I. Pavlidis and J. Levine, “Thermal image analysis for polygraph testing,” IEEE Eng. Med. Biol. Mag., vol. 21, no. 6, pp. 56–64, Nov./Dec. 2002. [16] D.U. Silverthorn, “Respiratory physiology,” in Human Physiology: An Integrated Approach, 2nd ed. Englewood Cliffs, NJ: Prentice-Hall, 2001, pp. 498–508. [17] R. Murthy, “Feasibility study of breathing rate computation using infrared imaging, M.S. thesis,” Dept. Comput. Sci., Univ. Houston, Dec. 2004. [18] N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Trans. Syst., Man, Cybern., vol. 9, no. 1, pp. 62–65, 1979. [19] G.C. Holst, Common Sense Approach to Thermal Imaging. Bellingham, WA: SPIE, 2000, p. 144. [20] M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis, and Machine Vision, 2nd ed. Pacific Grove, CA: Brooks/Cole, 2001. [21] I. Pavlidis, V. Morellas, P. Tsiamyrtzis, and S. Harp, “Urban surveillance systems: From the laboratory to the commercial world,” Proc. IEEE, vol. 89, no. 10, pp. 1478–1497, Oct. 2001. [22] J.T. Tou and R.C. Gonzalez, Pattern Recognition Principles. Reading, MA: Addison-Wesley, 1974. [23] A. Pednekar, I.A. Kakadiaris, and U. Kurkure, “Adaptive fuzzy connectedness-based medical image segmentation,” in Proc. 2002 Indian Conf. Computer Vision, Graphics, Image Processing (ICVGIP’02), Ahmedabad, India, pp. 457–462.

MAY/JUNE 2006

67

Study of Facial Skin and Aural Temperature

BY EDDIE Y.K. NG, WIRYANI MULJO, AND B. STEPHEN WONG

his article aims to study the correlations between the facial skin surface and contact aural temperatures using two types of infrared (IR) scanners: 1) imager with external temperature reference source (TRS) to set the threshold temperature (no actual reading but with color display mode only) and 2) imager with direct threshold temperature setting, which can be used for all type of operations (with skin temperature reading). To mimic the real situation at checkpoints such as air- and seaports, control screening and data analysis on 30 subjects with three known conditions for an intervenient time using a Type 2 scanner were investigated. The noncontact ear temperature was more closely related to contact-measured aural reading after running, drinking a hot beverage, and under normal conditions. Next, the results based on 750 blind sample sizes obtained by the Type 1 scanner are vital in determining two very important pieces of information: the best (yet practical) region on the face to analyze the suggested procedure to achieve an optimal preset threshold temperature for the Type 1 imager since that type is widely used. Human body temperature undergoes only slight variations in a course of a person’s lifetime. Variations of the body temperature may be due to fever, exertion, and extremely high or low environmental temperatures. A sudden infectious disease outbreak that began in mid-March 2003, popularly known as severe acute respiratory syndrome (SARS), and the avian flu are highly contagious and deadly diseases [1], [2]. A SARS-infected patient usually undergoes fever approximately ten days after the day of infection; similarly, the symptoms of bird flu are fever and cough. The outbreak has ignited studies and research (and even the general public interest) in the field of IR imaging system [5]. Based on Stefan-Boltzmann Law, all objects are continually emitting radiation at a rate and with a wavelength distribution that depends on the temperature of the object and its spectral emissivity, ε(λ) [6]–[8]. An IR scanner system is a device that captures IR energy that emits a short wavelength invisible to the human eye and converts it to electric voltage or current signal. The signal is then displayed in a pattern that is visible to the human eye [6]. In the field of IR thermography, blackbody is defined as an object that absorbs all radiation energy that impinges on it and, conversely, is a perfect radiator, ε = 1. In fact, real objects are all nonblackbody emitters due to the following three factors: energy absorptance, energy transmittance, and energy reflectance. These factors are highly dependent on the nature of the object, such as its temperature and emissivity. Several external parameters, such as transmittance of the air, ambient temperature, humidity, and distance also affect the accuracy of

T Using Infrared Imagers With and Without Temperature Reference Source

68 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Human body temperature undergoes only slight variations in a course of a person’s lifetime.

thermal camera. The emissivity of clean, dry human skin is close to the emissivity of blackbody, ε = 0.98. The actual value of the emissivity of particular area in a human face may depend on face contour, makeup (cosmetics), sweat, and skin complexion. Thermal imagers available in the market during the SARS outbreak in 2003 can be broadly divided into four different types [7], [13], [14]: 1) thermal imager system (TIS) without any temperature indication but with external TRS, 2) TIS with skin temperature indication and internal TRS, 3) TIS with skin temperature indication defined by external TRS, 4) TIS with body temperature indication defined by an external TRS and a temporal thermometer. TIS offers much promise to be used as a fast, blind mass-screening device to detect potential SARS or bird flu patients. Large effort has hitherto been put to examine the correlation between IR imaging temperature readings and deep body temperature values for the Type 2 imager [5], [6], whereas Types 3 and 4 were reported in the press but have yet to be used widely. Also, there are efforts to minimize the differences and to find alternative solutions to screen out suspected SARS or avian flu patients. This article aims to study the aforesaid correlations using the first two types of scanners. The data for patients entering a hospital would be very different than data for passengers walking through aerobridges in airports. Air motion, ambient temperature, perspiration (e.g., from hand luggage), glasses, physical strain, alcohol, a hot beverage, cosmetics, immediate past history of the face, or even deliberate cooling of the face are just few of such variables. Control screening and data analysis on subjects with three known conditions (well rested, after a hot drink, and after running) are thus investigated. Materials and Methods Apparatus ➤ Braun digital tympanic thermometers ➤ Mercury thermometers ➤ IRTRS (IR system operated at a hospital with three black-

body calibrators or TRS) [9] ➤ Set of computers ➤ Strip thermometers ➤ ThermalVision900 [10]

Materials ➤ Hot tea

Software ➤ ImageTool ➤ ImageJ ➤ Microsoft Excel

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Systems Description

The study was basically conducted by performing two major temperature measurements taken by two systems: 1) IRTRS with only color display mode and 2) ThermalVision900 (TVS900) with direct temperature readings. The IRTRS screens and captures color images and saves them as a black to white range of color displays. Darker colors represent lower temperatures, and lighter colors represent higher temperatures. The IRTRS used three blackbody calibrators, which serve as reference sources of the system. Each of the blackbodies was set to emit radiation at constant temperatures of 32 ◦ C, 35.1 ◦ C, and 37 ◦ C. The TVS900 on the other hand, has its own blackbody reference located within the scanner. The system is able to display and save color images. The system allows the object parameters setting to be varied [emissivity, ambient and atmosphere temperatures, object distance, transmittance, and relative humidity (RH)]. It is also able to measure the temperature of any points or areas of the image captured. Thus, it offers flexibility for the user to determine temperatures at various sites as compared to the IRTRS. The system comes together with a set of lenses for temperature taking at different distances. Procedures of Data Collection IRTRS (No Temperature Reading but with Color Display Only)

The system was set up in an indoor nonairconditioned passageway. Blackbody calibrators and a computer monitor (to store images) were set in line with a yellow box on the ground where the hospital staff and visitors stood when images were captured. The thermal camera itself was planted in the ceiling, and it was 4.5 m above the yellow box. Hospital staff were stopped, and their tympanic temperatures of both ears were taken. After that, they were asked to look straight ahead for their images to be captured. The samples were taken randomly (the subjects might be the same but at different times). Ambient temperature in each hour for every day was also recorded (29 ± 2◦ C; RH <80%). A total number of 750 blind data were collected within a week. TVS900 (With Direct Skin Temperature Reading)

The system was operated in Metrology Laboratory, Nanyang Technological University, a temperature-controlled environment (23 ± 1◦ C; RH <80%). There were 30 people that participated as the subjects of the study for three different conditions: normal, after consuming hot drinks, and after running. A 20◦ lens that is suitable for objects placed at a distance of more than 0.5 m was used. Throughout the temperature-taking session, the scanner was fixed at a distance of 1 m off the subjects. The parameters MAY/JUNE 2006

69

setting of the system were: emissivity = 0.98, atmosphere temperature = 30.5◦ , subject distance = 1 m, transmittance = 0.965, and RH = 0.8. Under normal conditions, body temperatures were taken in three different sites (ears, mouth, and temple) using ear digital (clinical Braun Thermoscan IRT 3520+), mercury, and strip thermometers, respectively. At almost the same time, IR images of front and both sides of the subject’s face were also being captured. Under the second condition, each person was to run the same distances of 7 × 15 m (some people run in the sun without perspiring much). Immediately after they completed the distance required, any sweat on their faces was wiped dry, and three face images (the front and both sides) were recorded for an intervenient time. At the same moment, temperature measurements using contacting devices (digital, mercury,

38 37.8

and strip thermometers) were also taken. The steps in temperature taking under the third condition were performed in a similar manner to that of the running case, except that people were asked to drink a cup of hot tea. The skin surface temperatures (forehead, adjacent to the eyes, temples, and ears) captured in the images were analyzed using the thermal scanner software [10]. The parameters of interest were the average temperature within a small circular pointer from a pointer to a point on specific sites. Results and Discussion

A data analysis performed on 750 blind samples showed that, in general, the left ear temperature is slightly higher (not > 1 ◦ C) than the right ear (Figure 1). Since the ear temperature resembles brain temperature, it makes sense to measure the left ear temperature for better correlation. The analysis also confirmed that the temperature of a healthy adult human is approximately 36.8 ◦ C, and the temperature of a feverish adult is 37.7 ◦ C or higher.

37.6

Temperature

37.4 37.2

Below 20 20–29 30–39

Data Analysis with IRTRS (Using External TRS and Color Display Only)

The IRTRS located in a hospital initially had its TRS set to 35.1 ◦ C. This value 36.8 50–59 followed the setting of the IRTRS operation in an air-conditioned environment. 36.6 Obviously, a different TRS setting 36.4 should be used for IRTRS operation in a nonair-conditioned environment. An 36.2 image of a typical feverish staff member 36 taken by IRTRS is shown in Figure 2 Left Right Left Right Left Right Left Right (light area of intensity over forehead). The analysis using ImageTool, ImageJ, Febrile Nonfebrile Febrile Nonfebrile and Microsoft Excel, however, resulted in a TRS of 35.5 ◦ C to obtain red spot Male Female areas that cover 30% of the total face and neck area. The procedure to obtain Fig. 1. Left versus right ear temperatures (◦ C, based on age group and sex). the TRS temperature threshold consisted of four steps: 1) Ten images of identified feverish staff members were analyzed to extract the three representative parameters for each Blackbody image. They were the maximum gray value of adjacent (35.1 °C - TRS) (inner corner) of eyes and the average gray values for both blackbody calibrators (as appeared in the upper-left images as two circles, lighter in color, Figures 2 and 3). We Blackbody extracted the parameters using ImageTool as follows: (37 °C - high) • The region of interest (ROI) in the image was chosen by selecting it with a square-shaped cursor. The cursor was placed around the eyes area to obtain the maxiBlackbody (32 °C - low) mum gray value of the adjacent eyes. In the case of obtaining average values of blackbody calibrators, a cursor of the same size was positioned in the middle Low Setting of the circles. The gray values of both blackbody calibrators captured by the scanner varied in values for every image (due to variation in the operating environment, quality of blackbodies, thermal imagers with different degrees of temperature drift between self corFig. 2. Image with a low threshold setting value (a feverish rection, uniformity within field of view, minimum staff). detectable temperature difference, error and stability 37

70 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

40–49

MAY/JUNE 2006

Since the ear temperature resembles brain temperature, it makes sense to measure the left ear temperature for better correlation.

% Area

of threshold temperature, distance effect, and detector sizes). were calculated to check whether they cover up to 30% of Theoretically, these variations should not exist since the blackthe face and neck area of the subject. This step was repeated body calibrators were expected to emit constant radiation with the “yes” answer until the desired threshold value was throughout the measurement. obtained. This value is being used in the operation of IRTRS 2) The temperature in Celsius scale (◦ C) was obtained by to help a layman to differentiate conveniently the feverish assuming a linear correlation between its values and the patients from the normal ones (see Table 1). Figure 3 illusgray values of both blackbody calibrators. As an example, trates a typical distribution of red spot area (in %) as a funcx1 and y1 correspond to a blackbody that was set to 32 ◦ C; tion of ear temperature. The correlation is rather low due to x3 and y3 correspond to a blackbody that was set to 37 ◦ C; the small sample size of available febrile cases and because and x2 and y2 correspond to the adjacent eye area. the manual extraction of numeric data is time-consuming for • Low gray value, x1 : 70.167 the current Type 1 scanner. • Low temperature, y1 : 32 ◦ C The highest temperature over the front area of face and • High gray value, x3 : 117.857 neck is the area adjacent to the eyes. This finding was • High temperature, y3 : 37 ◦ C obtained by setting a high threshold value for any images. • Maximum gray value, x2 : 129 The red spots that remained when a high threshold value • Tympanic temperature: 38.1 ◦ C was set indicated that the spots had the same temperatures • Maximum temperature, or higher than the corresponding threshold value. In fact, y2 =((x2 − x1 )/(x3 − x1 ))((y3 − y1 ) + y1 = 38.168 ◦ C there are quite a few arteries around the eye (the ophthalmic As expected, tympanic temperature resembled the deep artery is in vicinity to the lacrimal caruncle and it is conbody temperature taken by contact method, whereas y2 resembled the adjacent eye temperature taken by the nonArea Versus Temp contact method. 90 3) Initially, a gray value (x4 ), which is 80 also termed threshold, was randomly y = 59.784x - 2209.7 set to obtain red spot areas of 30% of 70 the overall face and neck area. Since R2 = 0.352 the blackbodies’ gray values were not 60 % Actual the same for every image, the temperaLinear (% Actual) 50 ture value (y4 ) for this specific threshold (x4 ) was calculated using the same 40 method as in Step 2. 4) The temperature value (y4 ) for every 30 image was set to the same value, and it 20 was converted to the respective gray 37.4 37.6 37.8 38 38.2 value again for different images. Several Temp images of feverish and nonfeverish subjects were observed by setting the value to the threshold calculated previously. Fig. 3. Area percentage (red spots) versus ear temperature distribution (◦ C, limitNext, the red spot areas that appeared ed febrile cases).

Table 1. Sample of determining a TRS value (iterative-steps are indicated as 1–6). Sample

30% of

% of red spot

Threshold temp

Number

Total Face Area

Face Area

Red Spot Area

over total

Threshold (x)

value (y) ◦ C

421

8560

1712

1815

21.2

119

1

35.5

2

613

5375

1075

2198

125.7

4

35.5

3

307

5727

1145.4

4659

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

5

40.9 81.4

6

103.5

35.5

MAY/JUNE 2006

71

clearly indicate the higher temperature adjacent to the eyes (not included here). Statistical results showed that eye temperatures obtained by IRTRS fluctuated even for the nonfebrile subjects. Although individual variations occur, the inside human body temperature is fairly constant, with small variations over the daily cycle and small variations between individuals. The correlation coeffiContact Versus Noncontact y = 0.8577x + 4.2134 39.000 cient between eye (noncontact) and ear 2 = 0.5409 R temperatures (contact) is about 0.541 38.000 Data Points for Non(Figure 4; note that many data points overfebrile and Febrile 37.000 lap operating points for normal subjects). Patients 36.000 The low correlation may be due to the Linear (Data Points for 35.000 Nonfebrile and Febrile instability of the camera in capturing Patients) images (proven by different gray values for 34.000 the blackbodies in every image), the 33.000 changes of the operator experience in tem35 36 37 38 39 perature screening, standing positions that Tympanic were not always the same (many of the subjects were in a hurry), the small number of Fig. 4. Linear regression of eye versus ear (tympanic contact) temperature (◦ C). the feverish population, and the method to Note that many data points overlap operating points for the normal subjects. analyze the data collected. The method to acquire gray values of the blackbodies as described previously involved time-consuming manual interpretation. Thus, accuracy in Contact Versus Noncontact obtaining the correct values was still greatly 37.500 dependent on the operator experiences. Figure 5 presents the distribution of contact Eye–Nonfebrile 36.500 (ear) and noncontact (eye) temperatures for Tympanic–Nonfebrile both normal and febrile subjects. The 35.500 Eye-Febrile Pearson Moment Correlation Coefficients (R values) of ear versus eye are 0.224 and Tympanic-Febrile 34.500 0.3 for normal and febrile subjects, respectively. These values are low, as expected, 33.500 since we are plotting them separately. 0 5 10 15 20 25 30 In reality, the body determines a temPerson perature as its so-called set point at any Fig. 5. Graph showing eye and ear temperatures (◦ C) of febrile and nonfebrile one time during the body temperature patients (R values are 0.224 and 0.3 for normal and febrile subjects, respectively). regulation [12]. Fever happens if the hypothalamus detects pyrogens and then raises the set point. The time course of a typical fever can be divided into three stages. When the Table 2. Statistical analysis of significant deviation and standard error. fever begins, the body attempts to raise its temNumbers/Site Temple (Left–Right) Cheek (Left–Right) perature, but vasoconstriction occurs to prevent heat loss through the skin. For this reason, some –0.02513 Mean 0.040106952 individuals at this stage of fever (at the rising Standard Error 0.028265764 0.02846 slope and immediately after the fever begins or Median 0 0 the falling slope after the fever breaks) will not be detected by the scanner if it is not designed Mode 0 0 to detect the subject at the plateau of the fever Standard Deviation 0.386528511 0.389181 (with her/his high core temperature). Temperature (°C)

Eye Noncontact

nected to the optic nerve) [11]. Thus, a small area of skin near the eyes and nose offers the body core temperature to be measured since the thin skin in this area has the highest amount of light energy, making it a preferred point. With a higher threshold setting, the same image as Figure 2 would

Sample Variance

0.14940429

0.151462

Kurtosis

1.086850727

2.057879

Skewness Range

–0.159101714

–0.30164

2.6

3

Minimum

–1.4

–1.5

Maximum

1.2

1.5

Sum

7.5

–4.7

Count

187

72 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

187

Data Analysis with TVS900 on Subjects with Various Conditions

Reproducibility of both the instrument and physiological assumptions was established by comparing paired left-right readings of the temples and cheeks (not shown here). It is observed that a majority of the data has a very small difference between the left and right readings. In fact, it can be concluded that zero difference was obtained in the highest number of patients. MAY/JUNE 2006

Contact Versus Noncontact (Hot Drink)

Contact Versus Noncontact (Normal)

37 36 35 34

0

10

20

30

Ear Contact (Left) Ear Contact (Right) Ear Noncontact (Left) Ear Noncontact (Right)

Temperature

Temperature

38

37.75 36.75 35.75 34.75 33.75 32.75

0

10

Contact Versus Noncontact (After Running) 37.5 Temperature

30

Person (b)

Person (a)

36.5 35.5 34.5 33.5 0

20

Ear Contact (Left) Ear Contact (Right) Ear Noncontact (Left) Ear Noncontact (Right)

10

20

30

Ear Contact (Left) Ear Contact (Right) Ear Noncontact (Left) Ear Noncontact (Right)

Person (c)

Fig. 6. (a) Normal condition. (b) After a hot drink. (c) After running.

The highest differences between the left and right readings were around ±0.9 and ±1.5◦ C for the temple and cheek, respectively. These differences are considered low since the number of occurrences with these variations is low. The statistical analysis of significant deviation and standard error from normal distribution is included in Table 2. Therefore, this TVS900 imager can be said to have considerable good reproducibility. The observation of 30 people’s images taken by TVS900 [10] revealed that the most stable and hottest site is the ear area, regardless of the external condition applied (in this case, after running and after consuming a glass of hot tea). Generally, there is not much difference between the left and the right temperatures of both ears and the adjacent eye area. Temperatures of the left and right side temples were slightly scattered for the three conditions. Temperature measurement should not be taken at the forehead since the observation showed that the data fluctuated highly (this observation also exists for eye temperature as in the previous section). Contact methods, such as mouth temperature measured by mercury thermometer, best resemble deep body temperature, but it is time-consuming and not a practical site during SARS outbreaks. Generally, this temperature is slightly higher than both ear temperatures taken by tympanic thermometer. The correlation of the ear temperature (contact versus noncontact) at subject conditions specified as normal (resting without extra activity = 0.614), after consuming a hot drink (0.714), and after running 105 m (without perspiring much = 0.787), respectively. Based on statistical results, the ear temperature measured by the contact tympanic therIEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

mometer better mimics the actual body temperature. The expected low correlation due to noncontact measurement further confirmed the dynamic balance of heat production; transfer and loss from the various sites of the head in living human beings is a complex process (more so with extra work done by the subject) in which physiologic mechanisms are continuously in action. More analysis with a large sample size is necessary. Figure 6 summarizes ear temperature readings under three different subject conditions. Conclusions

Accurately mapping the skin surface with body temperatures has been a very important issue for disease diagnostics. Arising from the SARS outbreak in 2003 and the urgent need to screen for febrile individuals through mass screening, IR thermography received more attention in the media worldwide than ever before. Suddenly, IR professionals were supposed to be the heroes that would save the world from this evil disease. Now that the dust has settled, one should ask, what was really the IR contribution? This is a critical and sensitive question; when one goes through some airports lately, it is not uncommon to find that they are still not using IR cameras correctly to detect fever (at least not technically correct, such as focusing issue and distance to spot size ratio >5 mm2 , though it was a political success as far as psychological “warfare” was concerned). Sadly, the general public is still rather uneducated about the most basic physics and methodologies involved in IR thermography. However, it is certain that one can perform a good screening of probable fever cases—if we use the right method and cameras that are thermally stabilized and calibrated. MAY/JUNE 2006

73

We have studied the correlation between the facial skin surface and measured tympanic membrane temperature in two types of IR imagers. The results showed that the TVS900 imager Type 2 has good capability in reproducing readings. The noncontact ear temperature was more closely related to contact measured aural (tympanic) reading after running, drinking a hot beverage, and normal conditions. However, the correlation between ear (contact) and eye (noncontact) temperatures taken by IRTRS was relatively lower than what was expected. It is thus essential to give more attention on the device calibration before performing the data collection. The procedures previously suggested allow for more systematic and scientific ways of detecting feverish persons suspected of SARS. The future development of a computerized data processing method is, however, preferable so that less manual work is involved. In the statistical analysis, a setting of of 35.5 ◦ C was found suitable for an IRTRS that is potentially workable. The results are interesting and will likely be helpful both in designing new IR imagers and in determining how to use them most effectively. The current research application will also remain of interest and be useful for reference by both local and overseas manufacturers of thermal scanners, users, and various government and private establishments. As the elevation of body temperature is a common presenting symptom for many illnesses (including infectious diseases such as SARS). thermal imagers are useful and essential tools for the mass screening of body temperature during public health crises where widespread transmission of infection is a concern at places such as hospitals and cross-border checkpoints.

Wiryani Muljo graduated from Nanyang Technological University, Singapore, in 2005. She completed the design specialization course from the School of Mechanical and Aerospace Engineering.

B. Stephen Wong is an associate professor and has been a lecturer in Nanyang Technological University, Singapore, for 18 years, mostly with the School of Mechanical and Aerospace Engineering, where he presents course modules and conducts research in nondestructive testing (NDT). Before this, he worked for 13 years in the United Kingdom, conducting research. He has a Ph.D. in physics from the University of Manchester, United Kingdom, and has published more than 40 papers in various journals and conferences, mainly in NDT. He is a member of the Overseas Advisory Panel for the British Institute of NDT for Singapore and is a technical NDT auditor for the Singapore Laboratory Accredition Service. Address for Correspondence: Eddie Y.K. Ng, School of Mechanical and Aerospace Engineering, College of Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798. Phone: +65 6790 4455. Fax: +65 6791 1859. E-mail: [email protected]. References

Acknowledgments

E.Y.K. Ng would like to thank Dr. W.M. Bai and Dr. L.S.J. Sim for helping the IR measurement in a hospital setting and E.C. Kee for plotting some of the graphs. He also wants to express his appreciation to members of the Ad-hoc Technical Reference Committee on Thermal Imagers under Medical Technology Standards Division by SPRING [13], [14], FLIR Systems, IRTRS Company, and a hospital of SingHealth Group, Singapore, for sharing their views and interests on “Thermal Imagers for Fever Screening— Selection, Usage and Testing.” Eddie Y.K. Ng graduated from Cambridge University in 1992. He is an associate professor at the Nanyang Technological University in the School of Mechanical and Aerospace Engineering, Singapore. He is an associate editor for the Journal of Mechanics in Medicine and Biology and the International Journal of Rotating Machinery and the Chinese Journal of Medicine and regional editor for the Computational Fluid Dynamics Journal. He has published more than 212 papers in refereed international journals, international conference proceedings, and others. His interest is in computational fluid dynamics, turbomachinery aerodynamics, nanoscale computation, thermal imaging, human physiology, bioinformatics, and biomedical engineering. 74 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

[1] S.M. Peiris, S.T. Lai, L.L.M. Poon, Y. Guan, L.Y.C. Yam, W. Lim, J. Nicholls, W.K.S. Yee, W.W. Yan, M.T. Cheung, V.C.C. Cheng, K.H. Chan, D.N.C. Tsang, R.W.H. Yung, T.K. Ng, and K.Y. Yuen, “Coronavirus as a possible cause of severe acute respiratory syndrome,” Lancet, vol. 361, no. 20, pp. 1319–1325, 2003. [2] T.G. Ksiazek, D. Erdman, C.S. Goldsmith, S.R. Zaki, T. Peret, S. Emery, S. Tong, C. Urbani, J.A. Comer, W. Lim, P.E. Rollin, S.F. Dowell, A.E. Ling, C.D. Humphrey, W.J. Shieh, J. Guarner, C.D. Paddock, P. Rota, B. Fields, J. DeRisi, J.Y. Yang, N. Cox, J.M. Hughes, J.W. LeDuc, W.J. Bellini, and L. J. Anderson, “A novel coronavirus associated with severe acute respiratory syndrome,” New England J. Med., vol. 348, no. 20, pp. 1953–1966, 2003. [3] E.Y.K. Ng and G.J.L. Kaw, “IR scanners as fever monitoring devices: Physics, physiology and clinical accuracy,” in Biomedical Engineering Handbook, Nicholas Diakides, Ed. CRC Press, 2006, pp. 24-1–24-20. [4] E.Y.K. Ng, “Is thermal scanner losing its bite in mass screening of fever due to SARS?,” Med. Phys., vol. 32, no. 1, pp. 93–97, 2005. [5] E.Y.K. Ng, G. Kaw, and W.M. Chang, “Analysis of IR thermal imager for mass blind fever screening,” Microvascular Res., vol. 68, no. 2, pp. 104–109, 2004. [6] E.Y.K. Ng and N.M. Sudharsan, “Numerical modelling in conjunction with thermography as an adjunct tool for breast tumour detection,” BMC Cancer, Medline J., vol. 4, no. 17, pp. 1–26, 2004. [7] S.G., Burnay, T.L. Williams, and C.H. Jones, Application of Thermal Imaging. England: Adam Hilger, 1988. [8] E.F.J. Ring and B. Phillips, Recent Advances in Medical Thermology. New York: Plenum, 1982. [9] IRTRS, private communications, Singapore, 2003. [10] FLIR Systems [Online]. Available: http://www.flir.com (accessed 15 Mar. 2005) [11] Virtual hospital, Atlas of Human Anatomy (Transl.: R.A. Bergman and A.K. Afifi) [Online]. Available: http://www.vh.org/adult/provider/anatomy/atlasofanatomy/ plate17/index.html (accessed 11 Mar. 2005) [12] A.C. Guyton, Textbook of Medical Physiology, 10th ed. Philadelphia, PA: Saunders, 2000. [13] Standards Technical Reference, Thermal Imagers for Human Temperature Screening Part 1: Requirements and Test Methods, TR 15-1, Spring Singapore, 2003. [14] Standards Technical Reference, Thermal Imagers for Human Temperature Screening Part 2: Users’ Implementation Guidelines, TR 15-2, Spring Singapore, 2004.

MAY/JUNE 2006

COMAR Reports Dennis W. Blick

report of COMAR activities 2005–2006

he past year has been relatively quiet for COMAR; many members have been very busy with standards-setting in the IEEE International Committee on Electromagnetic Safety (ICES; formerly IEEE SCC 28). No COMAR meeting was held in summer 2005 as many members chose not to travel to Dublin, Ireland, where the meeting would ordinarily have occurred in conjunction with meetings of IEEE/ICES and the Bioelectromagnetics Society. Meetings were held in December 2004 and December 2005 in San Antonio, Texas, where ongoing efforts to develop future COMAR Technical Information Statements (TISs) were discussed. A TIS on how IEEE standards for human exposure to electromagnetic fields are set was published in this mag-

T

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

azine in the March/April 2005 issue. This was intended to help clear up questions about the openness of the process and the possibility for input as well as to explain how the experts who take part in this process are chosen. Most recently, we published a TIS in Health Physics on electromagnetic exposure of medical personnel working with patients in magentic resonance imaging (MRI) machines. We felt the audience for this document— medical doctors and technicians— would be best reached by publishing in Health Physics rather than in this magazine. The document received expert input from Food and Drug Adminstration personnel as well as from magnetic resonance imaging device manufacturers. Current and ongoing projects include a TIS explaining how epidemiology

works and how to interpret epidemiologic findings. Reports of such studies looking at exposures associated with cellular telephones are frequently reported in the press and are liable to a great deal of misunderstanding. A draft addressing consumer exploitation (marketing of various highly questionable devices to shield or “protect” consumers from nonionizing electromagnetic fields) is currently in circulation within COMAR. COMAR members are beginning work on a TIS dealing with radiological terrorism. As always, we welcome input from members of EMBS, both suggestions for future TISs and help on those in progress. Please check our Web page: http://ewh.ieee.org/soc/embs/comar/ for a list of all available COMAR documents as well as a link to a listing of current COMAR members.

MAY/JUNE 2006

75

Patents

Maurice M. Klee

the U.S. patent that reached around the world

an a United States patent be infringed by activities in a foreign country? That was the question before the Court of Appeals for the Federal Circuit (CAFC) in the recent case of Union Carbide v. Shell. The case involved a patented method for making the industrial chemical ethylene oxide using silverbased catalysts. Union Carbide and Shell were major players in this market and had a history of fighting over patents on catalysts. Shell won the first battle in the 1980s, and Union Carbide struck back in 1999. The amounts of money involved were staggering since ethylene oxide (EO) is used in making everything from polyester resins to hair shampoo. Competition in the EO market revolves around price, and thus the most efficient producer has a distinct advantage. Efficiency, in turn, depends on the catalysts used to make EO. Prior to Union Carbide’s work, the highest efficiency that had been achieved was 65%. Union Carbide discovered that it could achieve higher efficiencies by combining silver with cesium and lithium. It patented the use of these catalysts to achieve these higher efficiencies and offered to license Shell, but Shell turned the offer down. A seven-year patent infringement suit then ensued. After two jury trials, Union Carbide won, and it won big. By the time the counting was over, Shell had to pay Union Carbide over US$100 million in damages. One might think that US$100 million was enough. But Union Carbide didn’t. In a preliminary ruling, the trial judge had said that Union Carbide was not entitled to any damages for Shell’s sales of catalysts to its foreign customers.

C

76 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Union Carbide felt that such damages were proper because the only use for Shell’s catalysts was to practice Union Carbide’s patented method. Both at trial and on appeal, Union Carbide based its claim on a section of the Patent Laws [Section 271(f)], which Congress had enacted in 1984 in response to another court fight about offshore infringement. That case— Deepsouth v. Laitram—involved shrimp deveining machines. Laitram and Deepsouth both made these machines and Laitram had the patents. It sued Deepsouth for U.S. and foreign infringement. The U.S. infringement was easily decided in Laitram’s favor, but the trial court refused to extend Laitram’s apparatus patent to machines built in the U.S. but assembled in Brazil. Ultimately, the Supreme Court had to decide the issue. It ruled that U.S. patents are limited to the territory of the United States and could not reach machines that only came into existence outside of the country. Congress eventually overturned the Deepsouth decision by enacting Section 271(f). This section, in relevant part, i.e., 271(f)(2), provides that it is an act of infringement to supply “any component of a patented invention” from the United States with the intent that the component will be employed (combined) outside the United States in a manner that would infringe if done in the United States, if the component is 1) especially adapted for use in the invention and 2) not a staple article or commodity of commerce suitable for a substantial noninfringing use. Before the Union Carbide case, Section 271(f)(2) had been limited to apparatus and composition patents. Union Carbide wanted to extend its scope to method patents. It argued

that Shell’s catalysts satisfied all of the requirements of Section 271(f)(2) and, in particular, qualified as a “component of a patented invention,” i.e., the key component of Union Carbide’s method. A divided CAFC agreed with Union Carbide. The majority could see no reason why Union Carbide should not be entitled to obtain further damages based on Shell’s offshore sales of catalysts. The court thus expanded the scope of the U.S. patent laws and held that Union Carbide’s method patent covered sales of Shell’s U.S.-made catalysts anywhere in the world. Shell may petition the Supreme Court to review the CAFC’s decision and, thus, there may be more developments on this issue. However that may turn out, the Union Carbide case is another example of the shrinking size of the world. Courts respond to changing realities, and with the global marketplace and outsourcing, one can expect that offshore infringement will become an increasingly common issue faced by patent owners and the infringers they sue. Dr. Maurice Klee practices patent, trademark and copyright law in Fairfield, Connecticut. He received a B.S. in physics from the University of Illinois, a Ph.D. in biomedical engineering from Case Western Reserve University, and a J.D. from George Washington University. He is a member of Phi Beta Kappa and Order of the Coif. He is a former assistant professor in the College of Engineering at Michigan State University and a former staff fellow at the National Institutes of Health. A copy of the full text of the Union Carbide v. Shell decision can be obtained from the CAFC’s Web site at http://www.fedcir.gov.

MAY/JUNE 2006

Emerging Technologies healthcare applications of RF identification Dorin Panescu

adio frequency identification (RFID) technology had long been described as the next big thing in high tech. In the simplest terms, RFID is a technology that utilizes RF for communication between a reader and a tag. The tag carries a unique identification number. This number is transmitted to the reader anytime the tag is queried. The unique ID number can then be referenced in a database for additional information regarding the tagged item [1]. Once used during World War II to distinguish returning English airplanes from inbound German ones, technological advances and decreased costs are now facilitating the use of RFID in a variety of public and private sector settings, from hospitals to highways [2], [3]. Perhaps the first work exploring RFID is the landmark 1948 paper by Harry Stockman [4]. Stockman predicted that ...considerable research and development work has to be done before the remaining basic problems in reflected-power communication are solved, and before the field of useful applications is explored. It required 30 years of advances in many different fields before RFID became a reality. Announcements made by large retail stores, such as

R

Wal-Mart [5], and by U.S. government agencies, such as the Food and Drug Administration (FDA) [6], have certainly contributed to a wider acceptance of, as well as debate over, the RFID concept. As a consequence, RFID manufacturers made significant progress towards miniaturizing the

RF Identification Technology

It required 30 years of advances in many different fields before RFID became a reality.

tags, making the systems user friendly and reducing costs. At the same time, consumer advocacy groups raised privacy concerns and asked legislatures to place limits on what information could be collected using RFID technologies [7]. While most development efforts were driven by improvements sought in supply chain management, notable milestones were reached in RFID use for healthcare applications. As one

Antenna 1. Tag Enters RF Field Created by the Antenna. 2. Antenna's RF Signal Activates the Tag. 3. Coupler Sends a Modulated Signal. 4. Tag Demodulates the Signal and Returns Its Data to the Reader. 5. Coupler Sends Data to the Computer. 6. Computer Transmits New Data Through the Coupler to the Tag. PC Fig. 1. A typical RFID system.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

such example, in October 2004, the FDA gave final approval to VeriChip Corporation to sell their VeriChip RFID tags for subcutaneous implantation in patients [8], [9]. The VeriChip provides a unique identification number that may be used to access a database containing the patient’s identity and health information.

Coupler

Tag

RFID is similar in concept to bar codes. While bar codes are optically scanned with a laser and contain product-identification alphanumeric strings, RFID tags also contain identification information but are read via RF. There are four main parts of a RFID system. The tag is affixed to the item being tracked. The reader, or coupler, is the device that reads the tag, or writes information to it. The third part of the system is the antenna. The fourth part consists of a database/computer system that manages the information provided by tags. The basic system components are shown in Figure 1 [10]. The reader, or coupler, with an antenna emits a command to the tag and waits for a response. The command can be in an addressed or nonaddressed mode. In other words, the reader can look for certain tags. Any corresponding tag in the vicinity of the reader detects the signal and uses the energy from it to wake up. If the tag is of passive type, it rectifies the RF signal and uses it to supply operating power to the internal circuits. Active tags use an internal battery to power their circuits and may also initiate the communication with the reader. The tag decodes the signal, ensures that it is valid, and replies to the reader. The reader demodulates the information received from the tag and sends it to the computer or database management system.

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

77

Emerging Technologies (continued)

The scanning of RFID tags does not require line-of-sight access and can be done within a range of up to 30 ft, or even longer, depending on the characteristic of the system. In addition, RFID can contain a lot more information than the typical bar code. The electronic product code (EPC) stored in commercial tags has a range from a mere 96 b to over a megabit or 125 kB. The typical universal product code, known as UPC and referring to the data of a bar code, contains between 12–30 alphanumeric characters or about 30 B max [11]. For example, in product inventory applications, the UPC has a field in the alpha-numeric sequence that identifies the manufacturer (e.g., Dell, Kodak, etc.) and a field that identifies the particular type product (e.g., computer, film, etc.). Due to the larger amount of data that an RFID tag can store, the EPC can have a unique ID for each individual product that rolls off the assembly line—much like a serial number for each and every unit. For all these reasons, RFID can provide a lot more functionality than bar codes. Figure 2 illustrates a technical block diagram of the communication between a reader and a passive tag [12]. The reader generates an electromagnetic field that is received by the tag antenna. The RF voltage induced in the tag RF resonant circuit (i.e., antenna and C1 ) is rectified and converted to dc by the

diode and capacitor C2 . This dc voltage stored on capacitor C2 is used to power the tag chip, which contains the memory that carries the ID information. The chip also includes the necessary electronics (e.g., clocks, counters, parallelserial conversion, drivers, etc.) to address the memory and drive an RF modulator that actuates the tag antenna to send the ID information back to the reader. In the example shown in Figure 2, a simplified modulator is included that consists of a MOSFET. The ID information is used to modulate the gate voltage of the MOSFET, which then energizes the resonant circuit formed by the antenna and capacitor C1 . The electromagnetic field sent out by the tag is received by the reader, filtered using a bandpass filter and demodulated. As presented in Figure 1, the output signal from the reader is then sent to a computerized database management system. The size of the tag integrated circuit (IC) can be very small (e.g., some passive tags are 0.4 mm on the side). However, the overall tag size is defined by the size of its antenna. Typically, tag antennas are shaped as loops. Tag substrates can be paper, PVC, or Teflon. The antenna can be made from copper, aluminum, or conductive ink or paint. Once the chip is applied, an overlay of PVC, paper, or epoxy is added to protect everything from the elements.

RFID tags are classified by their ability to read and write data. The standards body EPCglobal categorizes RFID tags and readers into six different classes, which are outlined below [11], [13], [14]: ➤ Class 0 tags are read-only tags and are preprogrammed by the manufacturer. These are the simplest form of tags. The data is an alphanumeric identification string. The factory determines the numbers and programs the tags. The factory then disables the memory from any further updates. They do not have an ID number but announce their presence when passing through an antenna field. Class 0 tags can also be used for electronic article surveillance (EAS) or antitheft devices. ➤ Class 1 is a WORM (write once read many) tag. They can be programmed by the manufacturer in the factory or by the user, but data are written to the tag one time. These tags are usually used for simple product identification or for access cards or tokens (e.g., office access badges). ➤ Class 2 is a read/write tag. Users have access to read and write into the tag memory. Class 2 tags typically have more memory and can be used as data loggers. For example, in a manufacturing line, the tag can be updated at every step of production.

Magnetic Field H

~

Ri

C1

Chip

C2

BP Binary Code Signal Demod Reader

Fig. 2. A block diagram of the principle of communication between a reader and a passive tag.

78 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

MAY/JUNE 2006

➤ Class 3 tags are also read/write but

can also contain onboard sensors. They contain more electronics to record data from its onboard sensor, like temperature, pressure, or motion. The sensor data is recorded into the tag memory. Such tags have internal power supplies but are semi-passive, meaning that they cannot initiate the communication process with a reader. ➤ Class 4 are read/write but have integrated transmitters. These are the most sophisticated kind of tags. They can communicate with other tags and devices without the presence of a reader. To run independently they must contain their own power source or battery. ➤ Class 5 represents the readers. A more commonly used classification categorizes tags considering how the communication process is initiated. Based on this criterion, there are three different classes [2]: ➤ Passive tags have no onboard power source and do not initiate communication. A reader must first query a passive tag, sending electromagnetic waves that form a magnetic field when they “couple” with the

antenna on the RFID tag. Currently, depending on the size of the antenna and the frequency, passive tags can be read, at least theoretically, from up to 30 ft away. However, realworld environmental factors, such as wind and interference from substances like water or metal, can reduce the actual read range for passive tags to 10 ft or less. Passive tags are already used for a wide array of applications, including building-access cards, mass transit tickets, and, increasingly, tracking consumer products through the supply chain. Depending on the sophistication of the chip, such as how much memory it has or its encryption capability, a passive tag currently costs between 20 cents and several dollars. ➤ Semipassive tags, like passive tags, do not initiate communication with readers, but they do have batteries. This onboard power is used to operate the circuitry on the chip, storing information from various sensors, such as ambient temperature. Semi-passive tags can be used to create “smart dust”—tiny wireless sensors that can monitor envi-

ronmental factors. A grocery chain might use smart dust to track energy use or a vineyard to measure incremental weather changes that could critically affect grapes. Devices using smart dust, also known as motes, currently cost about US$100 each but, in a few years, reportedly could drop to less than US$10 apiece. ➤ Active tags can initiate communication and typically have onboard power. They can communicate the longest distances—100 ft or more. Currently, active tags typically cost US$20 or more. A familiar application of active tags is for automatic toll payment systems, like E-ZPass, which allow cars bearing active tags to use express lanes that do not require drivers to stop and pay. RFID tags can also be differentiated by their operating frequency range. Table 1 presents RFID frequency and characteristics [11], [12]. Lower frequencies are better able to penetrate and read through materials. Higher carriers can support faster bit rates of the data with the microwave range achieving rates of up to 2 Mb/s.

Table 1. RFID frequency bands and characteristics. RF range 125–140 kHz

13.56 MHz

860–930 MHz

2.45 GHz

Benefits Frequency range well accepted Tags work well near metal or in absorbent media (e.g., higher water content like animals, patients) Low cost Low power requirements Frequency range well accepted Tags work well in most environments Less noise sensitive Longer read-range potential, up to 20 m Growing commercial use Longer read-range potential, up to 10 m High reading speeds

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Limitations Limited range, less than 3 m

Commercial Use Animal, patient ID, auto antitheft systems

Slow reading speeds Sensitive to noise

Does not work well near metal Limited range, less than 3 m

Does not work well in moist environments Complex system developments. More line-of-sight required

Tracking of library books, airline baggage, apparel, pallet/ container, building access control Shipment, truck tracking

Vehicle access control, Wi-Fi, Bluetooth

MAY/JUNE 2006

79

Emerging Technologies (continued)

Figure 3 shows details of commercial passive tags provided by Maxell [15]. The tag encasing incorporates a loop antenna with overall dimensions of 2.5 mm × 2.5 mm. The device operates at 13.56 MHz. Figure 4 illustrates ICs (Mu-chip), block diagram, and passive tags realized using technology from Hitachi. The IC is 0.4 mm × 0.4 mm, requires an antenna with an active length in the 5-cm range, and operates at 2.45 GHz.

Table 2 provides some of the specs for this technology [16]. As shown in the block diagram, the RF voltage pickedup at the two antenna-connection points is rectified and used to supply the electronics that process and store the ID information. The same RF voltage is used to generate an internal clock. The information that can be identified can be encoded using 62–88 b. The power rectifier block is also used to generate RF oscillations that

RFID Chip : 2.5 mm × 2.5 mm

(a)

(b)

Fig. 3. Maxell Corporation’s (a) commercial tag and (b) IC.

0.4 mm

10-b Counter

Power Rectifier

Power-on Reset

128-b Rom

Clock Extraction

Digital Circuit

0.4 mm

Decoder

Analog Circuit (a)

(b)

(c)

Fig. 4. The (a) Mu-chip, (b) its block diagram, and (c) tags with technology from Hitachi.

80 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

are modulated—amplitude shift keying (ASK)—by the ID information and sent back to the reader. ASK encodes digital zeros and ones by modulating the amplitude of the carrier frequency accordingly. These tags can work within a range of up to 4 m from the reader. While the tags shown above are suitable for applications such as inventory and asset tracking, and positive ID of genuine drugs (e.g., determent of counterfeit drugs), the VeriChip RFID tag can be implanted subcutaneously in patients [9]. Figure 5 displays a seethrough view of the VeriChip and the tag resting on a patient’s hand [9], [17]. The microchip and its antenna measure just 11.1 × 2.1 mm. The tubeshaped VeriChip includes a memory that holds 128 characters of information (equivalent to half a page of A4 size paper), an electromagnetic coil for transmitting data, and a tuning capacitor, all encapsulated within a siliconeand-glass enclosure. The passive RF unit, which operates at 125 kHz, is activated by moving a companydesigned reader within about a foot of the chip. Currently, the chip is being used to store primarily patient identification and a limited amount of other information. The expected lifetime of the device is 20 years. The database that is accessed through the readerInternet link provides more of the information specific to the patient. The chip is packaged in a needle, which is then sterilized. Using a syringe connected to the needle that carries the device and a topical anesthetic, the VeriChip is implanted in the upper part of a patient’s right arm. The procedure takes approximately two minutes. The chip is contained in a biobond material (a polypropylene sheet) that has adhering properties with human tissue and is used to keep the body from rejecting the implanted chip. This sheet also protects the individual against any electromagnetic corruption. The reader is also an important part of the RFID chain. There is less standardization in terms of reader types and formats. Most tag manufacturers proMAY/JUNE 2006

vide either their own readers or technical support for users to design or order appropriate readers. For example, Figure 6 shows the hand-held reader used as part of the VeriChip system. Healthcare Applications

In the healthcare industry, RFID tracking technologies are seen as having the potential to be widely used for asset and inventory tracking, patient tracking, and drug counterfeiting deterrence. For a variety of reasons, adoption of RFID technology by the healthcare industry has been sluggish because payback is less immediately visible than what most companies prefer. Although costs are decreasing, many companies are reluctant to invest in a technology not yet widely adopted. However, recent developments have made RFID applications more compelling. RFID tags costs dropped from US$1 in 2000 to US$.20 in 2004 and are expected to fall to US$.05 in 2006. In 2004, readers cost about US$1,000 but are expected to fall to only US$200 in 2006 [18]. Coupling the RFID technology with the EPC standardized requirements will provide the capability to locate and track items throughout the supply chain, allowing significantly more data to be attached to items at the pallet and case level. EPCGlobal and EAN International are writing specifications on the content for 96-b EPC tags [13]. New regulatory requirements mandate that important information accompany each drug throughout the supply chain. Using RFID will allow healthcare companies to capture required information such as drug name, dosage, container size, number of containers, lot/control numbers, etc. Tampered or adulterated products entering the healthcare supply chain is a growing concern. In light of the terrorist attacks and the anthrax-tainted mail found in the United States, both consumers and manufacturers are looking for ways to keep drugs safe. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Asset and Inventory Tracking Applications

RFID has strong application potential with medical device companies as it would facilitate the identification of units by serial number. Return of implants on consignment with hospitals can occur more than 50% of the time. RFID technology that improves visibility into returns could enable faster redeployment since the company would know sooner when an unused product could be returned. Surgical instruments and other devices must be properly cleaned and packaged between uses. Tags on the instruments and readers on the sterilization chambers and storage cabinets

can validate proper cleaning and help locate needed instruments. Various errors would be completely traceable from manufacture to use, and preventative maintenance on equipment could be more accurately tracked. Large amounts of inventory typically can be found in hospital operating rooms. Lack of visibility in the supply chain coupled with the unauthorized purchase of certain items often results in the proliferation of “unofficial” inventory that could be reduced by properly managing the materiel ordering process [19]. Utilizing active RFID, tags can be placed on those assets which are most valuable, whether due to cost or operational

Table 2. Basic specifications for the Hitachi RFID Mu-chip. Parameter Carrier frequency Frequency bandwidth Chip size Chip thickness Chip technology Metallization Memory capacity Data transfer rate Minimum operating voltage Antenna connection Input impedance Power supply method Capacitance of power supply Minimum required current Clock signal Reading distance Minimum level of input energy Memory power dissipation Operational temperature (also depends on inlet type) ROM ID number format

Modulation style

Value 2.45 GHz 2.400–2.4835 0.4 mm × 0.4 mm 0.06 mm–0.15 mm 0.18-µm CMOS Three-layer aluminium 128 b 12.5 kb/s 0.5 V Anisotropic conductive film 60
Rectified electromagnetic energy 100 pF 3 µA 100 KHz Up to 400 mm according to reader power 3.5 dBm 0.06 µA −20 ◦ C − 100 ◦ C

Header: 1–6 b Service id: 10–36 b (according to class A–D) Application Data: 62–88 b (according to service ID) EDC Code: 24 b ASK

MAY/JUNE 2006

81

Emerging Technologies (continued)

(a)

(b)

Fig. 5. (a) See-through view of the VeriChip RFID tag. (b) The VeriChip RFID tag device rests on a patient’s hand.

necessity. In a typical 200-bed hospital, that could mean approximately 3,000 assets [19]. The active RFID tags beacon on a periodic basis and, in some cases, may offer the ability to detect certain conditions such as movement, tamper, or particular environmental conditions through the use of sensors that can be incorporated into the tags. Active RFID tags are able to be read from distances ranging from 20–1,000 ft or more, based on the surrounding environment and the characteristics of the asset. For assets such as wheel chairs, it may be sufficient to simply know that they

are within the building or perhaps are on a particular floor as the primary goal may be to prevent them from leaving with patients. Other assets such as intravenous pumps or certain surgical equipment may require that the location be determined and narrowed to a particular area within the facility. With currently available RFID technologies, a solution can be implemented whereby a nurse or technician can quickly determine the location of an asset by utilizing any computer that is tied into the hospital network. In addition to locating assets, a variety of queries and reports can be run by anyone who is provided access to the system, whether onsite or remote. Patient-Tracking Applications

Fig. 6. The VeriChip RFID reader shown retrieving the 16-digit patient ID from the VeriChip tag implanted in a patient’s arm.

82 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

RFID has been used in the form of chip implants for many years in animals for tracking and identification purposes. VeriChip Corporation provides 80% of the animal tags in the United States. In 2002, New Jersey surgeon Dr. Richard Seeling injected two VeriChips into himself. He placed one chip in his left forearm and the other near the artificial hip in his right leg. Dr. Seeling was motivated to find a better ID modality after he saw firefighters at the World Trade Center in September 2001

writing their Social Security numbers on their forearms with Magic Markers. In October 2004, the FDA approved the VeriChip RFID tags for subcutaneous implantation in patients [8], [9]. The VeriChip system provides a unique identification number that may be used to access a database containing the patient’s identity and health information. Per VeriChip Corporation, the chips could be implanted in young children or in adults with Alzheimer’s disease to help hospital staff identify people who cannot identify themselves. The VeriChip implant costs approximately US$200. The VeriChip reader currently costs about US$700. In another step towards facilitating use of RFID technologies for patient tracking, the FDA is moving rapidly to issue regulation to determine how hospitals can use RFID systems to identify patients and/or permit relevant hospital staff to access medical records. Drug Counterfeiting Deterrence

Pharmaceutical companies, distributors, and hospitals need technology to deter drug counterfeiting. The World Health Organization estimates that between 5–8% of global pharmaceuticals are counterfeit. In some countries, the percentage of counterfeit drugs is significantly higher at between 25–40%. Thus, the pharmaceutical industry reports that it loses several billion U.S. dollars per year due to counterfeit drugs [18], [19]. Recognizing the seriousness of the problem, in July 2003, the FDA established a Counterfeit Drug Force having the goals to: ➤ prevent the introduction of counterfeit drugs and biologics in the U.S. distribution system ➤ facilitate the identification of counterfeit drugs and biologics ➤ minimize the exposure of consumers to counterfeit drugs and biologics ➤ avoid the addition of unnecessary costs to the prescription drug distribution system or unnecessary restrictions on lower-cost sources of drugs. MAY/JUNE 2006

The FDA reported that the number of counterfeit drug investigations increased to over 20 per year in early 2000, as compared to about five per year in the late 1990s [6]. The Counterfeit Drug Force concluded that “the adoption and use of RFID as standard track-and-trace technology, which is feasible in 2007, would provide better protection.” The timeline for RFID adoption released by the FDA is as follows [6], [19]: ➤ 2004: pilots and feasibility studies ➤ 2005: RFID deployed on pallets, cases, and packages of highrisk products ➤ 2006: RFID on most pallets and cases of high-risk products and on some pallets and cases of other products ➤ 2007: RFID on pallets and cases of all products and most packages. In January 2006, Pfizer Inc., the largest drug manufacturer in the world, introduced RFID technology to track drug shipments. In a bid to halt counterfeiting of Viagra shipped to pharmacies and wholesalers, the drug maker attached RFID tags to all packages of its anti-impotence drug. Pharmacists in the United States are now able to use RFID readers to verify that their package of Viagra is genuine. Privacy Concerns

An article about RFID technologies would be remiss if it did not mention the privacy concerns surrounding this topic. The use of RFID technology has engendered considerable controversy and even product boycotts by consumer privacy advocates such as CASPIAN Founder Katherine Albrecht, who refers to RFID tags as “spychips.” Perhaps, the statement made by Debra Bowen, California State Senator, at a 2003 hearing summarizes these concerns most eloquently: “How would you like it if, for instance, one day you realized your underwear was reporting on your whereabouts?” [7], [17]. The four

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

main privacy concerns regarding RFID are [17]: ➤ the purchaser of an item will not necessarily be aware of the presence of the tag or be able to remove it ➤ the tag can be read at a distance without the knowledge of the individual ➤ if a tagged item is paid for by credit card or in conjunction with use of a loyalty card, then it would be possible to tie the unique ID of that item to the identity of the purchaser ➤ The EPCglobal system of tags create, or are proposed to create, globally unique serial numbers for all products, even though this creates privacy problems and is completely unnecessary for most applications. Most concerns revolve around the fact that RFID tags affixed to products remain functional even after the products have been purchased and taken home and thus can be used for surveillance and other nefarious purposes unrelated to their supply chain inventory functions. As it pertains to the healthcare industry, the main concerns related to the use of RFID technologies stem from the need to protect patients’ medical histories and information. As concerns such as those listed above have not been fully addressed yet by applicable regulations and industry standards, states such as California voted to limit the use of RFID technologies [7]. Although healthcare companies are slowly adopting RFID, usage is expected to accelerate when RFID technology prices drop and companies and advocacy groups become more confident of the applications. RFID strong functionality can clearly improve operational efficiency. Over the next few years, FDA recommendations for tagging drugs and managing hospital inventory and patient information should have a ripple effect on the healthcare industry adoption of RFID technology.

References [1] A. Wyne, “Radio frequency identification,” Rollpack Corp. [Online]. Available: http://www. plasticbag. com/conferences/rfidwhitepaper.pdf [2] “Radiofrequency identification: Applications and implications for consumers,” Fed. Trade Comm., Mar. 2005 [Online]. Available: http://www.ftc. gov/os/2005/03/050308rfidrpt.pdf [3] “What is RFID?” [Online]. Available: http:// www.palowireless.com/rfid/whatisrfid.asp [4] H. Stockman, “Communication by means of reflected power,” Proc. IRE, pp. 1196–1204, Oct. 1948. [5] A. Gonsalves, “Wal-Mart takes first shipments of RFID-tagged products,” Information Week, Apr. 30, 2004 [Online]. Available: http://www.informationweek.com/story/showArticle.jhtml?articleID=19400083 [6] “Combating counterfeit drugs,” U.S. Food Drug Admin., Feb. 2004 [Online]. Available: http://www. fda.gov/oc/initiatives/counterfeit/report02_04.html [7] T. Kontzer, “California senate OKs bill to limit RFID use,” Inform. Week, Apr. 29, 2004 [Online]. Available: http://www.informationweek. com/story/showArticle.jhtml?articleID=19205578 [8] Implantable Radio Frequency Transponder System, 510(k) Premaket Notification Database, K033440, U.S. Food Drug Admin. [Online]. Available: http://www.accessdata.fda. gov/scripts/ cdrh/cfdocs/cfpmn/pmn.cfm?ID=13528 [9] VeriChip Corp. Web site [Online]. Available: http://www. verichipcorp.com/solutions.html [10] V. Chachra, “Experiences in implementing the VTLS RFID solution in a multi-vendor environment,” World Library and Information Congress: 69th IFLA General Conference and Council, Aug. 2003, Berlin [Online]. Available: http://www.ifla. org/IV/ifla69/papers/132e-Chachra.pdf [11] J. Kabachinski, “An introduction to RFID,” IT World, Mar./Apr. 2005 [Online]. Available: http:// www.aami.org/publications/BIT/2005/ITMA05.pdf [12] S.C. Lee and G-Y. Lu, “A random walk into the world of RFID,” SCMJ Wireless Technol. Workshop, June 26, 2004 [Online]. Available at: http://www. scmj.org/Events/2004%20Archive/20040626%20wir eless%20workshop/A%20Random%20Walk%20into %20the%20World%20of%20RFID.pdf [13] EPCglobal Web site [Online]. Available: http://www.epcglobalinc.org [14] “Passive, battery-assisted passive and active tags: A technical comparison,” Intelleflex Corp., 2005 [Online]. Available: http://www.intelleflex. com/pages/Technical.pdf [15] Maxell Corporation Web site [Online]. http://www.maxell-usa.com/rfid/rfid.htm [16] Hitachi, Ltd. Web site [Online]. http://www. hitachi-eu. com/mu/ [17] “RFID,” Wikipedia [Online]. Available: http://en. wikipedia.org/wiki/RFID [18] “RFID in healthcare—A panacea for the regulations and issues affecting the industry?” UPS Supply Chain Solutions, WP.SCS.HC.631, 2005 [Online]. Available : http://www.upsscs.com/solutions/white_ papers/wp_ RFID_in_healthcare.pdf [19] “RFID applications in the healthcare and pharmaceutical industries,” RadiantWave, LLC, Feb. 2005 [Online]. Available: http://www.radiantwave. com/whitepapers/healthWP.doc

MAY/JUNE 2006

83

Engineering in Genomics

Elizabeth M. Flood, Robert S. Kumar, Rashmi Shah, Quinlan Amos Jonathan D. Wren, Ralph V. Shohet, and Harold R. Garner

melatonin administration does not affect isoproterenol-induced LVH

eft ventricular hypertrophy (LVH), a common sequella of hypertension, is associated with increased incidence of sudden cardiac death [1]. Studies suggest that there is regression of LVH after significant decrease in blood pressure with commonly prescribed antihypertensive agents [2]. Angiotensin- converting enzyme inhibitors, calcium antagonists, diuretics, and beta-blockers have been found effective in reducing LVH [3]. IRIDESCENT is a computational program that can detect previously unknown relationships between objects (small molecules, phenotypes, etc.) in MEDLINE [4]. In a previously reported study, IRIDESCENT was used to predict previously unknown relationships between cardiac hypertrophy and small molecules included in the MEDLINE database, and a relationship between cardiac hypertrophy and chlorpromazine was predicted and confirmed [5]. In this search, IRIDESCENT also predicted a relationship between cardiac hypertrophy and melatonin (N-acetyl5-methoxy-tryptamine), a protein involved in circadian cycle regulation [6]. Noncomputational analysis of the literature revealed a substantial body of work discussing relationships between melatonin and various deleterious cardiac conditions. High doses of melatonin, equivalent to ∼700-fold the dosage recommended in “over-the-counter” melatonin, have been found to decrease mean arterial pressure and heart rate [7]. Recently, repeated nighttime melatonin administration at levels comparable to that recommended with over-the-counter melatonin tablets was found to lower blood pressure in humans, although no effect on heart rate was observed [8].

L

84 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Melatonin has been suggested as a potential cardiotherapeutic agent for these and other reasons [9], [10]. Melatonin has been found to modulate vascular smooth muscle tone [11]. Studies conducted to characterize the effect of melatonin on the cardiovascular system in mice and humans have shown that melatonin protects cardiomyocytes against ischemia-reperfusion injury and raises

In this study, one effect of extended hypertension, cardiac hypertrophy, was not affected—even by very large doses of melatonin. the fibrillatory threshold of cardiac myocardium [12]. Melatonin has also been shown to prevent doxorubicininduced cardiac dysfunction and apoptosis [13]. A proposed mechanism by which melatonin exerts its cardiovascular effects is modulation of autonomic output via its action on the suprachiasmatic nucleus (SCN). In the setting of ischemic-reperfusion injury, observed effects are often attributed to the antioxidant capability of melatonin [14], [15]. Several studies have demonstrated the ability of melatonin to limit ischemia-reperfusion injury, but confounding factors often hamper data interpretation [15]. It has recently been suggested that the antioxidant effects of melatonin have previously been overestimated [16],

and, consequently, alternative mechanisms for the action of melatonin should be considered. It is possible that melatonin might have a direct effect on cardiomyocytes. Isoproterenol induces cardiac hypertrophy by its interaction with the β1- and β2-adrenergic receptors [17]. β-adrenergic receptors couple to the Gs protein to activate adenylyl cyclase and increase cAMP production. This leads to protein kinase A phosphorylation of certain regulatory proteins and metabolic enzymes, causing increased cardiac contractility and accelerated relaxation [18]. Isoproterenol also induces melatonin expression via β -adrenergic receptors [19], but this increase in melatonin levels has been suggested to be a compensatory response since melatonin can blunt the effect of isoproterenol [20]. Melatonin was demonstrated to attenuate isoproterenol-induced protein kinase A overactivation in the brain [21]. In spontaneously hypertensive rats, which typically have greater plasma noradrenaline and an increased proportion of β 2-adrenoceptors in the heart and a decrease in the chronotropic and mean arterial pressure response to isoproterenol relative to Wistar-Kyoto rats, melatonin decreased mean arterial pressure and heart rate and restored plasma noradrenaline concentrations, the chronotropic response to isoproterenol, and the proportions of β 1/β 2-adrenoceptrs in the heart to the levels in Wistar-Kyoto rats [7]. Some actions of melatonin in regulation of circadian and seasonal physiology are mediated through the high affinity Gi protein-linked melatonin receptors MT1 and MT2, both of which have been found in cardiac tissue [22], [23]. Both melatonin

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

receptors mediate adenylyl cyclase inhibition. The MT1 melatonin receptor mediates phospholipase Cβ activation. The MT2 melatonin receptor inhibits the soluble guanylyl cyclase pathway. Furthermore, melatonin inhibits phosphorylation of the transcription factor cyclic AMP response binding protein. Melatonin has been shown to stimulate calmodulin phosphorylation by protein kinase C [24]. Calmodulin associates with calcineurin to activate it, resulting in phosphorylation of transcription factors of the nuclear factor of activated T-cells (NFAT) family [25]. Studies have shown that activation of the calcineurin/NFAT pathway is sufficient for the development of cardiac hypertrophy [26]. Cellular signaling studies in Xenopus laevis and Labrus bimaculatus suggest that melatonin stimulation of the Mel1c receptor via Gβγ activates phosphoinositide 3-kinase (PI3K) that, directly or indirectly via mitogen-activated protein kinase (MAPK), activates phosphodiesterase [27]. Other studies have shown that melatonin induces PI3K/insulin receptor subunit 1 (IRS-1) and IRS-1/SH2-containing phosphotyrosine phosphatase 2 associations and downstream AKT (protein kinase B) serine phosphorylation and MAPK phosphorylation, respectively [28]. Studies have shown that PI3K is activated in pressure overload hypertrophy in a Gβγ dependent fashion [29]. Melatonin might have a direct effect on heart tissue through these pathways. The computer program IRIDESCENT predicted a relationship between melatonin and cardiac hypertrophy. Manual review of the literature confirmed the value of exploring this prediction. The study reported herein measured the effect of melatonin on the heart in the setting of direct cardiac myocyte stimulation with isoproterenol, thus minimizing the role of autonomic output from the SCN and clarifying the effect of melatonin on LVH, an outcome of chronic hypertension. IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Methods

An analysis of the MEDLINE database with the program IRIDESCENT using the search term cardiac hypertrophy is described in Wren et al. [5]. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised 1996) and all procedures were approved by the University of Texas Southwestern Medical Center Institutional Animal Care and Use Committee.

The computer program IRIDESCENT predicted a relationship between melatonin and cardiac hypertrophy.

LVH was induced in eight-weekold, male C57BL/6 mice (Jackson laboratories, Bar Harbor, Maine) by administration of 20 mg/kg/day isoproterenol hemisulfate (Sigma, St. Louis, Missouri) for seven days [30]. Isoproterenol was administered via osmotic minipumps (Alzet model 1007D, Cupertino, California). Pumps were inserted into the infrascapular subcutaneous tissue of the a n i m a l during anesthesia induced by Avertin (Aldrich, Milwaukee, Wisconsin) −0.02 ml of a 1.25% (w/v) solution per gram of body weight. A dose-ranging experiment was conducted in mice in which LVH was induced by isoproterenol to determine what levels of melatonin were tolerated by mice. Two mice received 0.075 mg/kg, two mice received 0.225 mg/kg, and two mice received 0.75 mg/kg melatonin (Sigma, St. Louis, Missouri) in 1:11 ethanol:saline by

intraperitoneal (i.p.) injection once daily for six days [31]. The highest dose produced no obvious symptoms, and subsequent experiments were performed with 0.75 mg/kg. Twenty mice were administered isoproterenol (20 mg/kg/day) via osmotic minipump. Starting 24 h after pump insertion, half of the mice received 0.75 mg/kg melatonin in 1:11 ethanol: saline by i.p. injection once per day for six days. This dosage has been used in previous research [5] but is equivalent to ∼700-fold the dosage typically recommended for over-the-counter melatonin. Control animals received 1:11 ethanol:saline in a similar manner. The melatonin-treated and control mice were kept in separate cages, and food consumption was monitored. Mice were maintained on a 12 h:12 h lightdark cycle and received standard rodent chow (Harlan Teklad, Madison, Wisconsin) and water ad libitum. In addition, in ten mice, minipumps containing normal saline were inserted as described above. Finally, three mice received melatonin injections for 21 days and, starting on Day 14, isoproterenol via minipump. Seven days after minipump insertion, the mice were euthanized with CO2 gas and cervical dislocation, and the hearts were removed. Each heart was trimmed to remove any remaining blood vessels, washed with sterile saline, blotted dry, and weighed. The left leg of each mouse was placed in 2 ml of 1% (w/v) sodium dodecyl sulfate (Sigma, St. Louis, Missouri) containing 20 µg/ml proteinase K (Qiagen, Valencia, California). Following overnight incubation at 50 ◦ C with agitation, any remaining tissue was removed, and the left tibia was measured with digital calipers (World Precision Instruments, Sarasota, Florida). Indexed LV mass was assessed by calculating heart weight:body weight and heart weight:tibia length ratios. Statistical Analysis

All values are expressed as mean ± standard deviation. One-way ANOVA (analysis of variance) was used to evaluate differences among groups, and MAY/JUNE 2006

85

Engineering in Genomics (continued)

post hoc comparisons between groups were performed by Tukey’s procedure. Unpaired two-tailed t-tests were used to compare the effects of the treatments given to the different groups. The groups were assumed to have equal variances. Differences were considered to be statistically significant when P < 0.05. Results

Initially, average mouse weights were 23.37 ± 1.87 g in the group administered isoproterenol and melatonin (ISO + MEL group, n = 10), 22.99 ± 1.41 g in the group administered isoproterenol and saline (ISO + SAL group, n = 10), and 23.12 ± 2.43 g in the group administered saline only (control group, n = 8). After the scheduled course of injections, average mouse weights were 22.37 ± 2.14 g for the ISO + MEL group, 22.58 ± 1.55 g for the ISO + SAL group, and 22.73 ± 2.45 g for the control group. On average, the mice in the ISO + MEL group lost 4.20 ± 6.11% of their starting

body weight, those in the ISO + SAL group lost 1.77 ± 3.99%, and those in the control group lost −1.55 ± 5.19% of their starting weight. Average heart weights following seven days of treatment were 0.1443 ± 0.0166 g for the ISO + MEL group, 0.1466 ± 0.0324 g for the ISO + SAL group, and 0.1059 ± 0.0136 g for the control group. Average tibia lengths were 16.9 ± 0.5 mm for the ISO + MEL, 17.0 ± 0.5 mm for the ISO + SAL group, and 16.9 ± 0.1 for the control group. Average heart weight:body weight ratios were 0.0065 ± 0.0006 for the ISO + MEL group, 0.0065 ± 0.0011 for the ISO + SAL group, and 0.0047 ± 0.0002 for the control group (Figure 1). ANOVA revealed no significant difference between mice treated with isoproterenol only and mice treated with isoproterenol and melatonin (q = 0.0279). ANOVA revealed a significant difference between mice treated with saline only and mice treated with isoproterenol only or mice treated with isoproterenol and melatonin (q = 7.779 and q

0.012 Heart Weight (g) / Tibia Length (mm)

Heart Weight (g) / Body Weight (g)

0.008 0.007 0.006 0.005 0.004 0.003 0.002 0.001 0

0.01 0.008 0.006

Discussion

0.004 0.002 0

(a)

(b)

Fig. 1. Mouse heart weight normalized by mouse body weight (a) and by = mice treated with melatonin by i.p. injection mouse left tibia length (b). and isoproterenol via mini pump.
= mice treated with saline by i.p. injection = mice treated with saline via mini pump. and isoproterenol via mini pump. Error bars represent ± 1 standard deviation. Melatonin did not significantly affect isoproterenol-induced cardiac hypertrophy as assessed by heart weight:body weight ratio.

•

•

86 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

= 7.751, respectively). P values for two-tailed t-tests conducted on these results were 0.9871 for the ISO + MEL group versus the ISO + SAL group, 4.23 × 10−8 for the ISO + MEL group versus the control group, and 8.01 × 10−5 for the ISO + SAL group versus the control group. Average heart weight:tibia length ratios were 0.0085 ± 0.0009 for the ISO + MEL, 0.0086 ± 0.0018 for the ISO + SAL group, and 0.0063 ± 0.0008 for the control group (Figure 1). ANOVA revealed no significant difference between mice treated with isoproterenol only and mice treated with isoproterenol and melatonin (q = 0.360). ANOVA revealed a significant difference between mice treated with saline only and mice treated with isoproterenol only or mice treated with isoproterenol and melatonin (q = 10.085 and q = 9.725, respectively). P values for two-tailed t-tests conducted on these results were 0.897 for the ISO + MEL group versus the ISO + SAL group, 1.38 × 10−5 for the ISO + MEL group versus the control group, and 0.0015 for the ISO + SAL group versus the control group. Three mice received melatonin injections for 21 days and isoproterenol via minipump for seven days. The heart weight:body weight ratios of these mice were not significantly different from the average heart weight:body weight ratio of mice treated with isoproterenol only for seven days.

The average heart weight:body weight ratio was not significantly different for mice treated with melatonin and isoproterenol for one week versus mice treated with isoproterenol only, although significantly larger hearts were observed in these groups than in the control group in which mice were not administered isoproterenol. The same differences between groups were observed when heart weight was normalized by tibia length. Melatonin administration for three weeks did not affect development of cardiac hypertrophy as quantified by heart weight:body weight ratio. MAY/JUNE 2006

Isoproterenol is an agonist at β -adrenergic receptors, and it induces LVH by direct stimulation of cardiac myocytes. These results show that daily injections of high dosages of melatonin do not affect heart weight:body weight or heart weight:tibia length ratios in the setting of isoproterenol-induced hypertrophy, suggesting that the cardiovascular effects of melatonin observed by others are not due primarily to a direct effect of melatonin on cardiac myocytes but to another mechanism, possibly melatonin modulation of SCN output. Melatonin has no effect on isoproterenol-induced LVH, suggesting that its cardiovascular effects are not due to a direct effect on cardiac myocytes. Melatonin has been found to have a significant effect on hypertension; however, in this study, one effect of extended hypertension, cardiac hypertrophy, which can ultimately lead to cardiac failure, was not affected, even by very large doses of melatonin in a chemically induced hypertrophy mouse model. It has been suggested that the effect of melatonin on hypertension is comparable to common antihypertensive agents [6]. How many hypertension medications that have a modest but significant effect on hypertension have no effect on outcomes of hypertension such as LVH? Acknowledgments

This study was supported by National Heart Lung and Blood Institute grant HL-99-024 and by the P. O’B Montgomery Distinguished Chair in Developmental Biology. Address for Correspondence: Harold R. Garner, Ph.D., University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-8591 USA. Phone: +1 214 648 1661, Fax: +1 214 648 1666. E-mail: [email protected].

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

References [1] L.L. Tin, D.G. Beevers, and G.Y. Lip, “Hypertension, left ventricular hypertrophy, and sudden death,” Curr. Cardiol. Rep., vol. 4, pp. 449–457, 2002. [2] J. Diez, A. Gonzalez, B. Lopez, S. Ravassa, and M.A. Fortuno, “Effects of antihypertensive agents on the left ventricle: Clinical implications,” Am. J. Cardiovasc. Drugs, vol. 1, pp. 263–279, 2001. [3] F.H. Messerli, “Hypertension and sudden cardiac death,” Am. J. Hypertens., vol. 12, pp. 181S–188S, 1999. [4] J.D. Wren and H.R. Garner, “Shared relationship analysis: Ranking set cohesion and commonalities within a literature-derived relationship network,” Bioinformatics, vol. 20, pp. 191–198, 2004. [5] J.D. Wren, R. Bekeredjian, J.A. Stewart, R.V. Shohet, and H.R. Garner, “Knowledge discovery by automated identification and ranking of implicit relationships,” Bioinformatics, vol. 20, pp. 389–398, 2004. [6] R.Y. Moore, “Circadian rhythms: Basic neurobiology and clinical applications,” Annu. Rev. Med., vol. 48, pp. 253–266, 1997. [7] H. Girouard, C. Chulak, M. LeJossec, D. Lamontagne, and J. de Champlain, “Chronic antioxidant treatment improves sympathetic functions and beta-adrenergic pathway in the spontaneously hypertensive rats,” J. Hypertens, vol. 21, pp. 179–188, 2003. [8] F.A.J.L. Scheer, G.A. Van Montfrans, E.J.W. van Someren, G. Mairuhu, and R.M. Buijs, “Daily nighttime melatonin reduces blood pressure in male patients with essential hypertension,” Hypertension, vol. 43, pp. 192–197, 2004. [9] D.J. Duncker and P.D. Verdouw, “Has melatonin a future as a cardioprotective agent?,” Cardiovasc. Drugs Ther., vol. 15, pp. 205–207, 2001. [10] Z. Chen, C.C. Chua, J. Gao, R.C. Hamdy, and B.H. Chua, “Protective effect of melatonin on myocardial infarction,” Am. J. Physiol. Heart Circ. Physiol., vol. 284, pp. H1618–H1624, 2003. [11] M. Jonas, D. Garfinkel, N. Zisapel, M. Laudon, and E. Grossman, “Impaired nocturnal melatonin secretion in non-dipper hypertensive patients,” Blood Press, vol. 12, pp. 19–24, 2003. [12] C.M. Blatt, S.H. Rabinowitz, and B. Lown, “Central serotonergic agents raise the repetitive extrasystole threshold of the vulnerable period of the canine ventricular myocardium,” Circ. Res., vol. 44, pp. 723–730, 1979. [13] M.F. Xu, S. Ho, Z.M. Qian, and P.L. Tang, “Melatonin protects against cardiac toxicity of doxorubicin in rat,” J. Pineal. Res., vol. 31, pp. 301–307, 2001. [14] A. Kladna, H.Y. Aboul-Enein, and I. Kruk, “Enhancing effect of melatonin on chemiluminescence accompanying decomposition of hydrogen peroxide in the presence of copper,” Free Radic. Biol. Med., vol. 34, pp. 1544–1554, 2003. [15] S. Bertuglia, P.L. Marchiafava, and A. Colantuoni, “Melatonin prevents ischemia reperfusion injury in hamster cheek pouch microcirculation,” Cardiovasc. Res., vol. 31, pp. 947–952, 1996. [16] G. Fowler, M. Daroszewska, and K.U. Ingold, “Melatonin does not ‘directly scavenge hydrogen peroxide’: Demise of another myth,” Free Radic. Biol. Med., vol. 34, pp. 77–83, 2003. [17] M.O. Boluyt, X. Long, T. Eschenhagen, U. Mende, W. Schmitz, M.T. Crow, and E.G. Lakatta, “Isoproterenol infusion induces alterations in expres-

sion of hypertrophy-associated genes in rat heart,” Am. J. Physiol., vol. 269, pp. H638–H647, 1995. [18] M. Kuschel, Y.Y. Zhou, H.A. Spurgeon, S. Bartel, P. Karczewski, S.J. Zhang, E.G. Krause, E.G. Lakatta, and R.P. Xiao, “β2-adrenergic cAMP signaling is uncoupled from phosphorylation of cytoplasmic proteins in canine heart,” Circulation, vol. 99, pp. 2458–2465, 1999. [19] H. Enzminger, K. Witte, and B. Lemmer, “Altered melatonin production in TGR(mREN2)27 rats: On the regulation by adrenergic agonists, antagonists and angiotensin II in cultured pinealocytes,” J. Pineal. Res., vol. 31, pp. 256–263, 2001. [20] F. Zalatan, J.A. Krause, and D.E. Blask, “Inhibition of isoproterenol-induced lipolysis in rat inguinal adipocytes in vitro by physiological melatonin via a receptor-mediated mechanism,” Endocrinology, vol. 142, pp. 3783–3790, 2001. [21] D.L. Wang, Z.Q. Ling, F.Y. Cao, L.Q. Zhu, and J.Z. Wang, “Melatonin attenuates isoproterenolinduced protein kinase A overactivation and tau hyperphosphorylation in rat brain,” J. Pineal. Res., vol. 37, pp. 11–16, 2004. [22] C. Ekmekcioglu, T. Thalhammer, S. Humpeler, M.R. Mehrabi, H.D. Glogar, T. Holzenbein, O. Markovic, V.J. Leibetseder, G. Strauss-Blasche, and W. Marktl, “The melatonin receptor subtype MT2 is present in the human cardiovascular system,” J. Pineal. Res., vol. 35, pp. 40–44, 2003. [23] L. Naji, A. Carrillo-Vico, J.M. Guerrero, and J.R. Calvo, “Expression of membrane and nuclear melatonin receptors in mouse peripheral organs,” Life Sci., vol. 74, pp. 2227–2236, 2004. [24] E. Soto-Vega, I. Meza, G. Ramirez-Rodriguez, and G. Benitez-King, “Melatonin stimulates calmodulin phosphorylation by protein kinase C,” J. Pineal. Res., vol. 37, pp. 98–106, 2004. [25] J.D. Molkentin, J.R. Lu, C.L. Antos, B. Markham, J. Richardson, J. Robbins, S.R. Grant, and E.N. Olson, “A calcineurin-dependent transcriptional pathway for cardiac hypertrophy,” Cell., vol. 93, pp. 215–228, 1998. [26] N. Frey and E.N. Olson, “Cardiac hypertrophy: The good, the bad, and the ugly,” Annu. Rev. Physiol., vol. 65, pp. 45–79, 2003. [27] T.P. Andersson, H.N. Skold, and S.P. Svensson, “Phosphoinositide 3-kinase is involved in Xenopus and Labrus melanophore aggregation,” Cell. Signal, vol. 3, no. 15, pp. 1119–1127, 2003. [28] G.F. Anhe, L.C. Caperuto, M. Pereira-Da-Silva, L.C. Souza, AE. Hirata, L.A. Velloso, J. CipollaNeto, and C.R. Carvalho, “In vivo activation of insulin receptor tyrosine kinase by melatonin in the rat hypothalamus,” J. Neurochem, vol. 90, pp. 559–566, 2004. [29] S.V. Naga Prasad, G. Esposito, L. Mao, W.J. Koch, and H.A. Rockman, “Gbetagamma-dependent phosphoinositide 3-kinase activation in hearts with in vivo pressure overload hypertrophy,” J. Biol. Chem., vol. 275, pp. 4693–4698, 2000. [30] M.H. Soonpaa and L.J. Field, “Assessment of cardiomyocyte DNA synthesis during hypertrophy in adult mice,” Am. J. Physiol., vol. 266, pp. H1439–H1445, 1994. [31] A. Cavallo and M. Hassan, “Stability of melatonin in aqueous solution,” J. Pineal. Res., vol. 18, pp. 90–92, 1995.

MAY/JUNE 2006

87

Retrospectroscope did Wheatstone build a bridge? L.A. Geddes

veryone in electrical engineering, electronics, and physics associates the name of Wheatstone with the bridge circuit for measuring electrical resistance. But did he really invent the bridge, and, if not, who did? If not, why is Wheatstone’s name associated with the bridge, and what, if anything else, did Wheatstone do? Briefly, his career interests ranged from acoustics to optics to electricity. The following will trace the path of this most creative person.

E

Acoustics

Charles Wheatstone was born at Barnwood near Gloucester, England, on 6 February 1802. His father sold musical instruments and music. In 1806, the Wheatstone family moved to London and resided at 128 Pall Mall, where his father continued his music business and also taught the flute. Young Charles became interested in music and invented novel musical instruments. His first was a keyed flute that he called a Harmonique; his most famous was the Enchanted Lyre, made in 1821, which was a stringed instrument that was suspended by a

Perceived Object

Left View

Right View

Fig. 1. Wheatstone’s mirror stereoscope.

88 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

stout cord from the ceiling. When played, it sounded like a harp, piano, and a dulcimer. The sound seemed to surround all those in the room. The wooden ceiling that supported the device probably acted as a sounding board. He also invented a gas-jet organ and a concertina. While in London, Charles attended a private school, but he had no formal scientific training. His inquisitiveness propelled him to learn new things continually, including several foreign languages. In 1832, his father died, and Charles and his brother William continued the family music business. Later, the business expanded to make and sell many of Wheatstone’s electrical instruments. Optics

In 1834, at the age of 32, Charles Wheatstone was the first to show that by viewing two images of an object obtained from two different vantage points the observer could perceive the object in three dimensions. Wheatstone made sketches of solid objects for the left and right images. Figure 1 shows Wheatstone’s stereoscope, which consisted of two mirrors arranged so that the left eye views the left image and the right eye views the right image. A complete account of his observations on the physiology of vision appeared in 1852 in the Philosophical Transactions of the Royal Society [1]. By this time, Wheatstone’s experimental genius was recognized. In 1834, he was appointed professor of experimental philosophy at King’s College, London, where he lectured until 1840. Because he did not like public speaking, he discontinued lecturing. However he remained at King’s College as a researcher until his death; he never retired because he loved his work. He died of bronchitis

in Paris while attending a scientific meeting on 19 October 1875. It is important to note that Wheatstone created stereographic (left and right) images long before photography was available. His remarkable perception of the way three-dimensional images are perceived by the brain is a tribute to his analytical genius. Electrical Measurements

The need to measure resistance accurately arose with development of the electric telegraph by Samuel F.B. Morse in the United States. The concept came to him during a transatlantic crossing from France to New York in 1832. The concept was of an electromagnetically driven stylus that inscribed dots and dashes on a moving strip of paper driven by clockwork. It was not until the fall of 1832 that he constructed the first working model. The first public exhibition of the apparatus was on the 2 September 1837, on which occasion the transmission was successfully effected through one-third of a mile of wire. Immediately afterwards, the first experimental telegraph line was to link Washington and Baltimore. This line was constructed between 1843 and 1844 under an appropriation by the U.S. Congress. The line was completed in May 1844, and the first message was transmitted from Washington to Baltimore on 27 May. The original Morse telegraph is on display at the Baltimore train station, which is in the center of a railroad museum. It is worth visiting. At that time there were no electrical units; the volt, ampere, and ohm were established much later in 1881 by the International Electrical Congress. Long before then, there arose the need to define a standard value for resistance so that comparisons could be made among conductors of different metals.

0739-5175/06/$20.00©2006IEEE

MAY/JUNE 2006

Investigators in various countries adopted their own standards of resistance. Wheatstone’s standard resistor was 1 ft of copper wire weighing 100 grains, which corresponds to a diameter of 1.8 mm. The French unit was 1 km of iron wire 4 mm in diameter. The German (Siemens) unit was a column of mercury, 1 m long with a cross-sectional area of 1 mm2 at 0 ◦ C. George Simon Ohm in 1826 published his paper on the electric circuit that gave us what we now know as Ohm’s law, namely that the resistance of a conductor depends on its length, inversely with the cross-sectional area, and, most importantly, the resistivity of the material from which it is made [2]. He also showed that the current was proportional to the voltage and inversely proportional to the resistance. But little was known of the conducting properties of wires of different metals. There were serious difficulties in measuring resistance accurately owing to the fact that battery voltages were not constant; they decreased with the amount of current delivered and decreased with the passage of time. The way that Ohm solved this problem was ingenious and practical only for laboratory work. Instead of using a Voltaic battery (which had an inconstant potential) as the voltage source, Ohm used multiple thermojunctions, one group at 100 ◦ C (boiling water) and the other at 0 ◦ C (snow or ice in water), thereby creating a voltage source with a constant potential and a constant internal resistance. Ohm’s current indicator consisted of a magnetic compass needle suspended by a flattened wire, 5-in long, to which torsion could be applied. When the needle was deflected from its position of rest in the magnetic meridian by the current, it was brought back to its original position by torsion. The angle through which the torsion element was turned was measured in centesimal division of a scale. The restoring force measured by the scale was proportional to the current. In the following year, 1827, Ohm published his book entitled Die IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

Galvanische Kette, mathematisch Bearbeitet [3]. It contained a theoretical derivation of his law and became more widely known than his paper of 1826, which gave the experimental details. Ohm’s experiments were met with considerable skepticism, if not disbelief. However, some did not share this view. Poggendorff and Fechner in Germany, Lenz in Russia, Wheatstone in England, and Henry in the United States expressed their admiration for Ohm’s work. In 1841, the Royal Society of London awarded Ohm the Copley Medal, probably as the result of Wheatstone’s enthusiasm. This endorsement gave credibility to Ohm’s law. Soon after Morse demonstrated the practicality of the electric telegraph, it was recognized that the revenue to be derived from it depended on the telegraph line length and the speed of signaling. Low-resistance lines and a sensitive,

rapidly responding galvanometer were needed. The second requirement was met by Thomson (Lord Kelvin) who patented the reflecting galvanometer in 1858 [4]. It consisted of a minute bit of steel watch-spring (or two or three such bits), which he cemented to the back of a light silvered glass mirror, suspended within a wire coil (of many turns to provide high sensitivity), by a single fiber of cocoon silk. By directing a beam of light upon the mirror from a lamp, the beam, reflected by the mirror, fell upon a long white card marked with the divisions of a scale, which was shaded from daylight or located in a dark corner. On the arrival of an electric current, the suspended magnet turned to the right or left and deflected the spot of light to the right or left on the scale. The beam of light served as a long, weightless lever providing exquisite sensitiveness, magnifying the minutest movements of the

MAY/JUNE 2006

89

Retrospectroscope (continued)

mirror. Interestingly, at that time the dots were transmitted with one polarity, the dashes with the opposite polarity. Because telegraph lines were many miles in length, it was necessary to know the resistance of a specimen of the line wire very accurately so that the total line resistance could be predicted with accuracy. Knowing the resistance of different wire diameters and different metals was essential for the success of the electric telegraph. Wheatstone’s contribution to the success of the electric telegraph was the accurate measurement of the resistance of wires, which started in the 1830s. He first invented a variable resistance (rheostat) that consisted of two identical parallel-grooved cylinders that could be made to rotate synchronously in opposite directions. One cylinder was of metal and the other was of wood. Bare German silver wire wound on the metal cylinder could be wound on to the wooden cylinder by a crank, thereby varying the length of bare German silver wire. German silver contains no silver; it is an alloy of copper, zinc and nickel. Wheatstone described his measuring circuit in the Philosophical Transactions of the Royal Society (London) in 1843 [5]. He called it the Differential Resistance Measurer and stated Mr. Christie in his Experimental Determination of the Laws of Magneto-Electric Induction, printed in the Philosophical Transactions of the Royal Society for 1833, has described a differential arrangement of which the principle is the same as that on which the instruments described in this section have been devised. To Mr. Christie must, therefore, be attributed the first idea of this useful and accurate method of measuring resistances. The differential principle that Christie used employed a galvanometer with two identical, oppositely wound coils. When the current in the two coils was equal, there was no deflection of 90 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

the galvanometer needle. This device was used to measure resistance by passing current through the unknown resistance and one of the galvanometer coils and passing current through a known variable resistance in series with the other galvanometer coil. Both circuits were connected to the same battery. When the known variable resistance was adjusted to obtain no galvanometer deflection, its resistance was equal to that of the unknown resistance. In reality, Christie’s circuit is a currentcomparison device that eliminated the problem of battery-voltage variation during the measurement. Wheatstone’s use of the differential principle was much different, and it did not require the use of a differential galvanometer; any sensitive galvanometer would suffice. He described his resistance-measuring circuit in 1843; he was 41. Like Christie’s circuit, Wheatstone’s Differential Resistance Measurer was independent of changes of battery voltage during measurement. However, with Christie’s method, resolution was poor when measuring low resistances. This difficulty was overcome by Wheatstone’s Differential Resistance Measurer. Wheatstone described his Differential Resistance Measurer in the Bakerian Lecture published in the Philosophical Transactions of the Royal Society in 1843. In part, Wheatstone stated [Figure 9.1] represents a board on which are placed four copper wires, Zb, Za, Ca, Cb, the extremities of which are fixed to brass binding-screws. The binding-screws Z, C are for the purpose of receiving wires proceeding from the two poles of a rheomotor, and those marked a, b are for holding the ends of the wire of a galvanometer. By this arrangement a wire from each pole of the rheomotor proceeds to each end of the galvanometer wire and if the four wires be of equal length and thickness, and of the same material, perfect equilibrium is established, so that a rheomotor, however powerful,

a z

c cd

b m

ef

Fig. 2. Wheatstone’s original bridge circuit, which he called a Differential Resistance Measurer.

will not produce the least deviation of the needle of the galvanometer from zero. (A rheomotor is a source of current. Figure 9.1 in his lecture is Figure 2 in this article.) As described above, Wheatstone’s Differential Resistance Measurer was a voltage-comparison circuit. Why it is called a bridge is not known. Some thought that the term derived from the galvanometer bridging two points of equal potential at balance. During the Victorian era, when Wheatstone lived, steel bridges were being built over rivers, and the diamond shape of the circuit— along with the vertical line added for the galvanometer connection—may resemble a bridge trestle. Finally if we call Christie’s circuit a current comparator, Wheatstone’s circuit is a voltage comparator. Its use to measure an unknown resistance required a voltage source, a galvanometer, a calibrated (known) resistor, and two equal resistances of any convenient value, easily made by winding equal lengths of wire of the same diameter on two wooden rods. References [1] C. Wheatstone, “Contributions to the physiology of vision—Part the second. On some remarkable, and hitherto unobserved, phenomena of binocular vision.” Phil. Trans. Royal Soc. London, vol. 142, pp. 1–17, 1852. [2] G.S. Ohm, “Bestimmung das Gesetzes, nach welchem metalle die cotelectricitat leiten Schweigger’s,” J. Chem. Phys., vol. 46, pp. 137–166, 1826. [3] G.S. Ohm, Die Galvanische Kette. Berlin: T.H. Riemann, 1827. [4] W. Thomson. “Reflecting galvanometer 1858,” U.K. Patent 329. [5] C. Wheatstone, “An account of several new instruments and processes for determining the constants of a Voltaic circuit.” Phil. Trans. Royal Soc. London, vol. 133, pp. 303–375, 1843.

MAY/JUNE 2006

Cellular/Tissue Engineering Maria Papadaki

tissue engineering and Eucomed news in brief

Vascularized Skeletal Muscle Tissue

o date, most approaches to vascularize engineered tissues have relied on implantation on the host animal. However, such approaches have not been successful on thick, highly vascularized tissues such as skeletal muscle. Levenberg et al. in their latest paper, which was published in the July 2005 issue of Nature Biotechnology (vol. 23, no. 7, pp. 879–882), presented an approach to form and stabilize endothelial vessel networks in vitro in three-dimensional (3-D) engineered skeletal muscle tissue. They hypothesized that vascularization in vitro could maintain cell viability, induce structural organization, and promote vascularization upon implantation. They used a 3-D multiculture system consisting of myoblasts, embryonic fibroblasts, and endothelial cells coseeded on highly porous, biodegradable spongelike polymer scaffolds (composed of 50% polylactic and 50% polyglycolic acid, with pore sizes of 225–500 µm). The addition of embryonic fibroblasts together with myoblasts and endothelial cells strongly promoted vascularization of the engineered muscle as evidenced by increases in the total area of endothelial cells and the number of endothelial lumens, compared with constructs seeded with myoblasts and endothelial cells only. The in vitro vessel-like structures were stable as evidenced by the large vessels present in the onemonth-old cultures and the twofold increase in the number of endothelial structures as compared with two-weekold tricultures. Furthermore, in vivo experiments showed that prevascularization of the implants improved implant vascularization and survival. This study emphasizes the importance of multicellular cultures in providing

T

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

appropriate signals for vascular organization of skeletal tissues. The mechanical properties of the engineered tissue, although not mentioned in this study, would certainly been improved if the engineered tissues had been grown under the appropriate mechanical stimulation in vitro. Other investigators, such as R. Dennis from the University of Michigan, are working on the promotion of excitability and contractility of engineered muscle constructs by providing nerve-derived trophic factors and synaptic stimulation. Due to the importance of the field, the National Institutes of Health has set up a Musculoskeletal Tissue Engineering Study Section to review grant applications for tissue engineering and related implants and device development projects that focus on the replacement or repair of damaged, missing, or poorly functioning musculoskeletal tissues, including bone, skeletal muscle, cartilage, tendon, and ligament.

What Is the Role of Eucomed?

Eucomed, the European Medical Technology Industry Association, represents directly and indirectly more than 3,500 business entities in Europe and beyond. Members include national trade and pan-European product associations, internationally active manufacturers of all medical technology products, smaller companies, and startups, such as those for tissue engineering or new materials. The Eucomed mission is to create and maintain a suitable framework by enabling the manufacturers of medical technology and other related economic actors to operate, innovate, and market in an open and competitive marketplace, with the ultimate objective to improve the access of patients and clinicians to modern and reliable

medical technology products. The medical technology industry is an important economic player in Europe, with some 8,500 enterprises (80% of which are small-to-medium employers) employing 400,000 people, most of whom are highly qualified, and a total market value of over ¤55 billion (30% of the global market). Medical technology is characterized by a constant flow of innovation, achieved by a high level of research and development combined with close cooperation with the end users. Research and development spending in the medical technology industry in Europe is generally recognized to be between 5–10% of sales, with some countries performing better than others. The key challenge areas are cardiovascular diseases, cancer, diabetes, and musculoskeletal conditions. A recent example of the role of Eucomed on the tissue engineering industry is illustrated in its response to the proposal of a regulation launched by the European Commission on 4 May 2005. The proposal covers human-tissue-engineered products, bringing them under the scope of medicinal products. Eucomed’s position is that since human-tissue-engineering products are not medicinal products, they cannot be regulated by the existing, unchanged rules for medicinal products. Existing requirements for clinical investigation and for good manufacturing practices for medicinal products should be appropriately adapted from the technical point of view, while maintaining their general ethical requirements. Eucomed has also put forward suggestions to address the specificity of combination products, in particular hybrid products made of human-tissue-engineered products and medical devices. For more information about Eucomed, visit http://www.eucomed.org. MAY/JUNE 2006

91

GOLD

Lisa Lazareck and Gerald Anleitner

EMBS goes for the GOLD!

f you are an IEEE member who graduated with your first professional degree within the last ten years, you are automatically part of IEEE GOLD! “Graduates of the Last Decade” is an IEEE entity working at providing benefits for young IEEE members after their student member status has expired. Around the world, there are over 47,000 GOLD members and 100 GOLD Affinity Groups. Typical GOLD activities include networking, educational activities, skill development workshops, and leadership seminars—events dedicated to making the transition from student to professional a smooth one. Fostering professional development is one of the key areas of interest for the IEEE and Engineering in Medicine and Biology Society (EMBS). (The mission statement is available at http://www. embs.org.) Of equal importance are the attraction of new members, retention of existing members, and facilitation of membership transition. Ten years ago, IEEE’s Regional Activities Board (RAB) created the GOLD program to do just that—focus volunteer efforts on the difficult transition period. The continuing goal of GOLD is to find out what students need from their Society at this particular stage of their careers and how their Society can, in turn, offer additional value of membership. One of the newest GOLD initiatives is the 2005 GOLD/Society Interaction Project, where the quality and quantity of interaction between GOLD and several selected pilot Societies is improved and augmented by increased communication and collaboration. The project’s overall goal is to increase attractiveness of Society membership and satisfaction with Society membership As one of the selected pilot Societies, EMBS is pleased to report that Lisa Lazareck (2004–2005 Student Representative) is volunteering as the 2006 EMB Society GOLD coordinator. Lisa

I

92 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

will dedicate her time to establish what exactly transitioning student members need, want, and must have from EMBS. Initial projects include advertising GOLD-related activities, leadership opportunities, etc. in IEEE Engineering in Medicine and Biology Magazine (see this column in the upcoming September/October 2006 issue of this magazine), on the student Web site (http://www.ewh.ieee.org/soc/embs/ student/), and through the EMBSGOLD mailing list; reporting to the EMBS Administrative and Executive Committees (AdCom, ExCom) and the IEEE GOLD Committee; and running a GOLD get-together at the EMBS annual conference (the first installment for the 28th Annual International Conference in New York). Get the most out of your developing career, find your place in the world of technology, and connect with colleagues across the globe through the GOLD-EMBS Interaction. “GOLDen” ideas, questions, concerns, opinions, desired activities, announcements, and offers to help are needed and most welcome by your new coordinator. Please email [email protected]—she would love to hear from you!

Discount for Graduating Student Members

Gerald Anleitner

Lisa Lazareck

IEEE student members who graduate and are elevated to full IEEE membership will automatically receive a one-year discount of 50% off of the full higher grade IEEE and Society membership dues rates upon renewal (see http://www.ieee.org/ organizations/rab/gold/). The offer is available once to IEEE student members upon their graduation and elevation to full higher IEEE membership. The discount is available to all IEEE student members graduating with an undergraduate or graduate degree. IEEE student members who previously graduated with a bachelor’s degree and received a discount upon their elevation to full IEEE member grade would not be eligible to receive another 50% discount if they returned to school and completed an advanced degree program. Lisa Lazareck is studying for a doctorate of philosophy in electrical engineering in the Engineering Science Department at the University of Oxford. Gerald Anleitner is currently working in the logistics area of a German car manufacturing company. He holds a degree in computer science and linguistics.

MAY/JUNE 2006

Conference Calendar

15–17 MAY 2006

30 AUGUST – 2 SEPTEMBER 2006

European Study Group of Cardiovascular Oscillations Jena, Germany Contact: Andreas Voss Phone: +49 3641 938958 Fax: +49 3641 938952 Web: http://www.esgco2006.de E-mail: [email protected]

28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society New York, New York Contact: EMB Executive Office Phone: +1 732 981 3451 Fax: +1 732 465 6435 E-mail: [email protected]

21–25 JUNE 2006

2006 Summer ASME Bioengineering Conference Amelia Island, Florida Contact: B. Barry Lieber Phone: +1 305 284 2330 Fax: +1 305 284 6494 Web: http://divisions.asme. org-/bed/events/summer06.html E-mail: [email protected]

American College of Clinical Engineering CCE Review Course Washington DC Contact: Arof Sibjam Phone: (818) 734-8384 Fax: (818) 886-0674 Email: [email protected]

Fourth International Conference On Ethical Issues In Biomedical Engineering Marriott New York At The Brooklyn Bridge

Contact: Subrata Saha Email: [email protected] 2–5 MAY 2007

6th IFAC Symposium on Modelling and Control in Biomedical Systems Reims, France Contact: Prof. Janan Zaytoon Phone: +33 3 26 91 32 26 Fax: +33 3 26 91 31 06 Web: http://www.univ-reims.fr/mcbms06 E-mail: [email protected]

3rd IEEE EMBS Special Topic Conference on Neural Engineering Kohala Coast, Hawaii Contact: Prof. Metin Akay Phone: +1 603 646 2230 Fax: +1 603 646 3856 E-mail: [email protected]

3rd Aegean “Biologie Perspective” Santorini Conference Santorini Island, Greece Contact: Brigitte Hiegel, Conference Secretariat Phone: +33 3 83 68 21 71 Fax: +33 3 83 32 13 22 Web: http://biol.prospectiveconf.u-nancy.fr E-mail: [email protected]

27 AUGUST – 1 SEPTEMBER 2006

26–28 OCTOBER 2006

World Congress on Medical Physics and Biomedical Engineering 2006 Seoul, Korea Contact: Sun I. Kim Web: http://www.wc2006-seoul.org/ E-mail: [email protected]

International Special Topics Conference on Information Technology in Biomedicine Congress Center Du Lac, Ioannina-Epirus, Greece Preconference Workshops: 24–25 October Contact: Lamprina Dimolika

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE

20 – 22 APRIL 2007

20–22 SEPTEMBER 2006

28 SEPTEMBER – 2 OCTOBER 2006 25 JUNE 2006

Phone: 302651098820 E-mail: [email protected]

22–27 AUGUST 2007

29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society Lyon, France Contact: EMBS Executive Office Phone: +1 732 981 3451 Fax: +1 732 465 6435 E-mail: [email protected]

13–18 SEPTEMBER 2009

World Congress on Medical Physics and Biomedical Engineering 2009 Munich, Germany Contact: Prof. Dr. Olaf Dössel Phone: +49 (0)721 608-2650 Fax: +49 (0)721 608-2789 Web: http://www.wc2009.de E-mail: [email protected]

MAY/JUNE 2006

93

ADVERTISERS’ INDEX The Advertisers’ Index contained in this issue is compiled as a service to our readers and advertisers: the publisher is not liable for errors or omissions although every effort is made to ensure its accuracy. Be sure to let our advertisers know you found them through IEEE Engineering in Medicine and Biology Magazine.

Advertiser

Page

URL

Phone

Ad Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 . . . . . . . . . . .www.adinstruments.com . . . . . . . . . . . . .+1 888 965 6040 Biopac Systems,Inc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .CVR 4 . . . . . . . . . . .www.biopac.com . . . . . . . . . . . . . . . . . . .+1 805 685 0066 CleveMed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 . . . . . . . . . . .www.CleveMed.com . . . . . . . . . . . . . . . .+1 877 CleveMed IEEE Enterprise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .CVR3 . . . . . . . . . . .www.ieee.org/try_enterprise . . . . . . . . . . .+1 732 981 0060 MathWorks,Inc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .CVR2 . . . . . . . . . . .www.mathworks.com . . . . . . . . . . . . . . . .+1 508 647 7040 NFS/NIH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89 . . . . . . . . . . .www.nsf.gov//eng/cts/workshops . . . . . . .+1 703 292 8371 Recruitment

Coming in July Clinical Neuroengineering Call your local advertising sales representative today! Space reservations: Material deadline:

6/2 6/7

Susan E. Schneiderman Business Development Manager [email protected]

+1 732 562 3946 Tel.; +1 732 981 1855 Fax.

ADVERTISING SALES OFFICES IEEE MAGAZINES 445 Hoes Lane, P. O. Box 1331 Piscataway NJ 08855-1331 www.ieee.org/publications/magazines James A. Vick Staff Director, Advertising Businesses +1 212 419 7767; Fax: +1 212 419 7589 [email protected] Susan E. Schneiderman Business Development Manager +1 732 562 3946; Fax: +1 732 981 1855 [email protected] Marion Delaney Advertising Sales Director +1 415 863 4717; Fax + 1 415 863 4717 [email protected] Product Advertising Midatlantic Lisa Rinaldo +1 732 772 0160; Fax: +1 732 772 0164 [email protected] NY, NJ, PA, DE, MD, DC, KY, WV Connecticut Stan Greenfield +1 203 938 2418; Fax: +1 203 938 3211 [email protected] CT New England/Eastern Canada Jody Estabrook +1 978 244 0192; Fax: +1 978 244 0103 [email protected] ME, VT, NH, MA, RI Canada: Nova Scotia, Prince Edward Island, Newfoundland, New Brunswick, Quebec

Southeast Bill Holland +1 770 435 6549; Fax: +1 770 435 0243 [email protected] VA, NC, SC, GA, FL, AL, MS, TN Midwest/Central Canada Dave Jones +1 708 442 5633 Fax: +1 708 442 7620 [email protected] IL, IA, KS, MN, MO, NE, ND, SD, WI, Canada: Manitoba, Saskatchewan Ohio Joe DiNardo +1 440 248 2456 Fax: +1 440 248 2594 [email protected] OH Midwest/Ontario, Canada Will Hamilton +1 269 381 2156 Fax: +1 269 381 2556 [email protected] IN, MI Canada: Ontario Southwest Josh Mayer +1 972 423 5507 Fax: +1 972 423 6858 [email protected] AR, LA, TX, OK Southern California/ Mountain States Marshall Rubin +1 818 888 2407 Fax: +1 818 888 4907 [email protected] AZ, CO,HI, NM, NV,UT, CA 93400 & below

Northern California/ Northwest/Western Canada Peter D. Scott +1 415 421 7950 Fax: +1 415 398 4156 [email protected] AK, ID, MT, WY, OR, WA, CA 93401 & above. Canada: British Columbia, Alberta Europe/Africa/Middle East Heleen Vodegel +44 1875 825 700; Fax: +44 1875 825 701 [email protected] Europe, Africa, Middle East Asia/Far East/Pacific Rim Susan Schneiderman +1 723 562 3946; Fax: +1 723 981 1855 [email protected] Asia, Far East, Pacific Rim, Australia, New Zealand Recruitment Advertising Midatlantic Lisa Rinaldo +1 732 772 0160 Fax: +1 732 772 0164 [email protected] NY, NJ, CT, PA, DE, MD, DC, KY, WV

New England/Eastern Canada John Restchack +1 212 419 7578; Fax: ++1 212 419 7589 [email protected] ME, VT, NH, MA, RI Canada: Quebec Nova Scotia, Prince Edward Island, Newfoundland, New Brunswick Southeast Thomas Flynn +1 770 645 2944 Fax: +1 770 993 4423 [email protected] VA, NC, SC, GA, FL, AL, MS, TN Midwest/Texas/Central Canada Darcy Giovingo +1 847 498 4520 Fax: +1 847 498 5911 [email protected] AR, IL, IN, IA, KS, LA, MI, MN, MO, NE, ND, SD, OH, OK, TX, WI, Canada: Ontario, Manitoba Saskatchewan,Alberta West Coast/Southwest/ Mountain States/Far East Tim Matteson +1 310 836 4064; Fax: +1 310 836 4067 [email protected] AK, AZ, CA, CO, HI, ID, MT, NM, NV, OR, UT, WA, WY Canada: British Columbia Europe/Africa/Middle East Heleen Vodegel +44 1875 825 700; Fax: +44 1875 825 701 [email protected] Europe, Africa, Middle East

3rd IEEE-EMBS International Summer School and Symposium on Medical Devices and Biosensors (ISSS-MDBS) “Wearable Body Sensor Networks and E-Textile Solutions for M-Healthcare” Summer School: Sept. 4-5, 2006

Symposium: Sept. 5-6, 2006

Venue: Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA, USA Conference Chair: Dr. Bonato, Paolo Harvard Medical School and Harvard-MIT Division of Health Sciences and Technology, USA

Conference Co-Chair: TBN Technical Program Chair: Dr. Zhang, Yuan-ting The Chinese University of Hong Kong, HKSAR, China

Technical Program Co-Chair: Dr. Asada, Harry Massachusetts Institute of Technology, USA

Technical Program Members: Dr. De Rossi, Danilo

Announcement Following the success of the international summer schools at University of WisconsinMadison and the Chinese University of Hong Kong, the 3rd ISSS-MDBS, sponsored by IEEE Engineering in Medicine and Biology Society, will be held at MIT, USA from Sept. 4-5, 2006. Innovative technologies in MDBS have been recently introduced to allow continuous monitoring of patient status in the home and community settings. The school will offer tutorial presentations and hands-on experience to familiarize attendees with emerging solutions in body sensor networks and e-textile for wearable medical devices. In addition, the combined symposium on Sept. 5-6, 2006 will provide participants with opportunities for presenting their work (papers are solicited) and interacting with other individuals with an interest in the field. By attending the three-day meeting, researchers and professionals will gain knowledge on the state-of-the-art of biosensors and e-textile solutions for wearable devices and their application in m-healthcare.

Highlighted Lectures 1. Body Sensor Networks: During the last decade, we have witnessed rapid

2.

University of Pisa, Italy

Dr. Dittmar, Andre CNRS/INSA Lyon, France

Dr. Jayaraman, Sundaresan Georgia Institute of Technology, USA

Dr. Korhonen, Ilkka VTT Information Technology, Finland

Dr. Lymberis, Andreas

developments in BSN aimed to gather data unobtrusively from wearable biomedical sensors. BSN technologies combine research advances in the academic and private sectors to monitor the physiological status of individuals over extensive periods of time. E-Textile for Wearable Medical Devices: Emerging e-textile technologies have gained the interest of academic researchers and professionals in the private sector. Recent work in this field has led to new e-textile materials and new ways to integrate etextile solutions in medical devices. New applications are emerging in m-healthcare.

Topics and Lecturers: 1. 2. 3. 4. 5. 6.

Sensing technology Prof John Webster, Univ of Wisconsin-Madison Body sensor networks Prof Joe Paradiso, Massachusetts Institute of Technology Wireless communication Prof Matt Welsh, Harvard University E-textile medical devices Prof Danilo De Rossi, University of Pisa E-textile systems Prof Harry Asada, Massachusetts Institute of Technology Wearable devices for M-health Prof Y.T. Zhang, The Chinese University of Hong Kong Home monitoring Prof Stephen Intille, Massachusetts Institute of Technology Applications in rehabilitation Prof Paolo Bonato, Harvard Medical School

Information Society & Media Directorate, European Commission

7. 8.

Dr. McAdams, Eric

Who should Participate

University of Ulster, N. Ireland

1.

Junior faculty members, researchers, and students with an interest in biomedical engineering and healthcare delivery. Researchers and engineers in the private sector with interest in the latest emerging technological developments in medical devices and their clinical applications.

Publicity Chair: Dr. Chen, Fei

2.

Local Arrangement Chair: TBN

Important Dates

Webmaster: Dr. Yan, Yong-sheng

Symposium Paper Submission Deadline: June 2, 2006 Notification of Paper Acceptance: June 16, 2006 Registration Deadline: June 30, 2006

For more information, please contact Dr. P. Bonato ([email protected]) Tel: +1 617 573-2745, Fax: +1 617 573-2769 Dr. Y.T. Zhang ([email protected]) Tel: + 852 2609-8459, Fax: + 852 2603-5558

WWW Information Updates:

http://bme.ee.cuhk.edu.hk/isss-mdbs/

IEEE is IEEE publishes information for the advancement of science — not for profit. As an IEEE member, Sanjay wouldn’t have it any other way. Sanjay wants his research to improve lives around the world. He relies on IEEE for the latest science and technology standards. And thanks to IEEE editors and peer reviewers, Sanjay can trust the information he finds. Every month, millions of researchers depend on IEEE for the latest results in their fields. With 126 journals and magazines, 600 annual conferences and more than 900 active standards, IEEE publishes the science that helps make the world a better place. To Sanjay, IEEE is for people, not profit. Discover what IEEE can be for you.

Go here.

Download the Free White Paper

“IEEE and Patents” by 1790 Analytics, LLC

www.ieee.org/discover

............................................. IEEE INFORMATION DRIVING INNOVATION