TLFeBOOK Biosignal and Biomedical Image Processing MATLA B-Based Applications JOHN L. SEMMLOW Robert Wood Johnson Medical School New Brunswick, New Jersey, U.S.A. Rutgers University Piscataway, New Jersey, U.S.A. MARCEL '.p% MARCELD EKKERIN, C. NEWY ORK BASEL DEKKER TLFeBOOK Althoughgreatcarehasbeentakentoprovideaccurateandcurrentinformation,neither theauthor(s)northepublisher,noranyoneelseassociatedwiththispublication,shallbe liable for any loss, damage, or liability directly or indirectly caused or alleged to be caused by this book. The material contained herein is not intended to provide specific adviceorrecommendationsforanyspecificsituation. Trademark notice: Product or corporate names may be trademarks or registered trade- marksandareusedonlyforidentificationandexplanationwithoutintenttoinfringe. LibraryofCongressCataloging-in-PublicationData AcatalogrecordforthisbookisavailablefromtheLibraryofCongress. ISBN:0–8247-4803–4 Thisbookisprintedonacid-freepaper. Headquarters MarcelDekker,Inc.,270MadisonAvenue,NewYork,NY10016,U.S.A. tel:212-696-9000;fax:212-685-4540 DistributionandCustomerService MarcelDekker,Inc.,CimarronRoad,Monticello,NewYork12701,U.S.A. tel:800-228-1160;fax:845-796-1772 EasternHemisphereDistribution MarcelDekkerAG,Hutgasse4,Postfach812,CH-4001Basel,Switzerland tel:41-61-260-6300;fax:41-61-260-6333 WorldWideWeb http://www.dekker.com The publisher offers discounts on this book when ordered in bulk quantities. For more information, write to Special Sales/Professional Marketing at the headquarters address above. Copyright2004byMarcelDekker,Inc.AllRightsReserved. Neitherthisbooknoranypartmaybereproducedortransmittedinanyformorbyany means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from thepublisher. Currentprinting (lastdigit): 10987654321 PRINTEDINTHEUNITEDSTATESOFAMERICA TLFeBOOK To Lawrence Stark, M.D., who has shown me the many possibilities... TLFeBOOK TLFeBOOK Series Introduction Over the past 50 years, digital signal processing has evolved as a major engi- neering discipline. The fields of signal processing have grown from the origin of fast Fourier transform and digital filter design to statistical spectral analysis and array processing, image, audio, and multimedia processing, and shaped de- velopments in high-performance VLSI signal processor design. Indeed, there arefewfieldsthatenjoysomanyapplications—signalprocessingiseverywhere in our lives. When one uses a cellular phone, the voice is compressed, coded, and modulated using signal processing techniques. As a cruise missile winds along hillsides searching for the target, the signal processor is busy processing the imagestakenalongtheway.WhenwearewatchingamovieinHDTV,millions ofaudioandvideodataarebeingsenttoourhomesandreceivedwithunbeliev- able fidelity. When scientists compare DNA samples, fast pattern recognition techniquesarebeingused.Onandon,onecanseetheimpactofsignalprocess- ing in almost every engineering and scientific discipline. Because of the immense importance of signal processing and the fast- growing demands of business and industry, this series on signal processing serves to report up-to-date developments and advances in the field. The topics of interest include but are not limited to the following: • Signal theory and analysis • Statistical signal processing • Speech and audio processing v TLFeBOOK vi • Image and video processing • Multimedia signal processing and technology • Signal processing for communications • Signal processing architectures and VLSI design Wehopethisserieswillprovidetheinterestedaudiencewithhigh-quality, state-of-the-artsignalprocessingliteraturethroughresearchmonographs,edited books, and rigorously written textbooks by experts in their fields. TLFeBOOK Preface Signal processing can be broadly defined as the application of analog or digital techniques to improve the utility of a data stream. In biomedical engineering applications, improved utility usually means the data provide better diagnostic information.Analogtechniquesareappliedtoadatastreamembodiedasatime- varying electrical signal while in the digital domain the data are represented as an array of numbers. This array could be the digital representation of a time- varying signal, or an image. This text deals exclusively with signal processing ofdigitaldata,althoughChapter1brieflydescribesanalogprocessescommonly found in medical devices. This text should be of interest to a broad spectrum of engineers, but it is written specifically for biomedical engineers (also known as bioengineers). Althoughtheapplicationsaredifferent,thesignalprocessingmethodologyused bybiomedicalengineersisidenticaltothatusedbyotherengineerssuchelectri- cal and communications engineers. The major difference for biomedical engi- neersisinthelevelofunderstandingrequiredforappropriateuseofthistechnol- ogy. An electrical engineer may be required to expand or modify signal processing tools, while for biomedical engineers, signal processing techniques are tools to be used. For the biomedical engineer, a detailed understanding of the underlying theory, while always of value, may not be essential. Moreover, considering the broad range of knowledge required to be effective in this field, encompassingbothmedicalandengineeringdomains,anin-depthunderstanding of all of the useful technology is not realistic. It is important is to know what vii TLFeBOOK viii Preface toolsareavailable,haveagoodunderstandingofwhattheydo (ifnothowthey do it), be aware of the most likely pitfalls and misapplications, and know how to implement these tools given available software packages. The basic concept ofthistextisthat,justasthecardiologist canbenefitfromanoscilloscope-type displayoftheECGwithoutadeepunderstandingofelectronics,soabiomedical engineer can benefit from advanced signal processing tools without always un- derstanding the details of the underlying mathematics. As a reflection of this philosophy, most of the concepts covered in this textarepresentedintwosections.Thefirstpartprovidesabroad,generalunder- standing of the approach sufficient to allow intelligent application of the con- cepts.Thesecondpartdescribeshow thesetoolscanbeimplementedandrelies primarily on the MATLAB software package and several of its toolboxes. This text is written for a single-semester course combining signal and image processing. Classroom experience using notes from this text indicates that this ambitious objective is possible for most graduate formats, although eliminating a few topics may be desirable. For example, some of the introduc- tory or basic material covered in Chapters 1 and 2 could be skipped or treated lightly for students with the appropriate prerequisites. In addition, topics such asadvancedspectralmethods (Chapter 5),time-frequencyanalysis (Chapter6), wavelets (Chapter 7), advanced filters (Chapter 8), and multivariate analysis (Chapter 9) are pedagogically independent and can be covered as desired with- out affecting the other material. Althoughmuchofthematerialcoveredherewillbenewtomoststudents, the book is not intended as an “introductory” text since the goal is to provide a working knowledge of the topics presented without the need for additional course work. The challenge of covering a broad range of topics at a useful, working depth is motivated by current trends in biomedical engineering educa- tion, particularly at the graduate level where a comprehensive education must beattainedwithaminimumnumberofcourses.Thishasledtothedevelopment of“core”coursestobetakenbyallstudents.Thistextwaswrittenforjustsuch a core course in the Graduate Program of Biomedical Engineering at Rutgers University. It is also quite suitable for an upper-level undergraduate course and would be of value for students in other disciplines who would benefit from a working knowledge of signal and image processing. It would not be possible to cover such a broad spectrum of material to a depth that enables productive applicationwithout heavy reliance on MATLAB- based examples and problems. In this regard, the text assumes the student has some knowledge of MATLAB programming and has available the basic MATLABsoftwarepackageincludingtheSignalProcessingandImageProcess- ingToolboxes. (MATLABalsoproducesaWaveletToolbox,butthesectionon wavelets is written soas not to require this toolbox,primarily to keep the num- ber of required toolboxes to a minimum.) The problems are an essential part of TLFeBOOK Preface ix this text and often provide a discovery-like experience regarding the associated topic. A few peripheral topics are introduced only though the problems. The codeusedforallexamplesisprovidedintheCDaccompanyingthistext.Since many of the problems are extensions or modifications of examples given in the chapter, some of the coding time can be reduced by starting with the code of a related example. The CD also includes support routines and data files used in the examples and problems. Finally, the CD contains the code used to generate many of the figures. For instructors, there is a CD available that contains the problem solutions and Powerpoint presentations from each of the chapters. Thesepresentationsincludefigures,equations,andtextslidesrelatedtochapter. Presentations can be modified by the instructor as desired. In addition to heavy reliance on MATLAB problems and examples, this text makes extensive use of simulated data. Except for the section on image processing, examples involving biological signals are rarely used. In my view, examplesusingbiologicalsignalsprovidemotivation,buttheyarenotgenerally very instructive. Given the wide range of material to be presented at a working depth, emphasisis placed on learningthe tools ofsignal processing; motivation is left to the reader (or the instructor). Organizationofthetextisstraightforward.Chapters1through4arefairly basic.Chapter1coverstopicsrelatedtoanalogsignalprocessinganddataacqui- sition while Chapter 2 includes topics that are basic to all aspects of signal and image processing. Chapters 3 and 4 cover classical spectral analysis and basic digital filtering, topics fundamental to any signal processing course. Advanced spectral methods, covered in Chapter 5, are important due to their widespread use in biomedical engineering. Chapter 6 and the first part of Chapter 7 cover topicsrelatedtospectralanalysiswhenthesignal’sspectrumisvaryingintime, aconditionoftenfoundinbiologicalsignals.Chapter7alsocoversbothcontin- uous and discrete wavelets, another popular technique used in the analysis of biomedical signals. Chapters 8 and 9 feature advanced topics. In Chapter 8, optimal and adaptive filters are covered, the latter’s inclusion is also motivated by the time-varying nature of many biological signals. Chapter 9 introduces multivariate techniques, specifically principal component analysis and indepen- dent component analysis, two analysis approaches that are experiencing rapid growth with regard to biomedical applications. The last four chapters cover image processing, with the first of these, Chapter 10, covering the conventions used by MATLAB’s Imaging Processing Toolbox. Image processing is a vast area and the material covered here is limited primarily to areas associated with medicalimaging:imageacquisition(Chapter13);imagefiltering,enhancement, andtransformation(Chapter11);andsegmentation,andregistration(Chapter12). Many of the chapters cover topics that can be adequately covered only in a book dedicated solely to these topics. In this sense, every chapter represents aseriouscompromisewithrespecttocomprehensivecoverageoftheassociated TLFeBOOK