Medical Image Processing: Techniques and Applications PDF
Preview Medical Image Processing: Techniques and Applications
BIOLOGICAL AND MEDICAL PHYSICS, BIOMEDICAL ENGINEERING BIOLOGICAL AND MEDICAL PHYSICS, BIOMEDICAL ENGINEERING Thefieldsofbiological andmedical physicsandbiomedical engineering arebroad, multidisciplinary anddynamic.Theylieatthecrossroadsoffrontierresearchinphysics,biology,chemistry,andmedicine. TheBiologicalandMedicalPhysics,BiomedicalEngineeringSeriesisintendedtobecomprehensive, coveringabroadrangeoftopicsimportanttothestudyofthephysical,chemicalandbiologicalsciences. Itsgoalistoprovidescientistsandengineerswithtextbooks,monographs,andreferenceworkstoaddress thegrowingneedforinformation. Books in the series emphasize established and emergent areas of science including molecular, membrane, and mathematical biophysics; photosynthetic energy harvesting and conversion; informa- tion processing; physical principles of genetics; sensory communications; automata networks, neural networks, andcellular automata. Equally important willbecoverage ofapplied aspects ofbiological andmedicalphysicsandbiomedicalengineeringsuchasmolecularelectroniccomponentsanddevices, biosensors,medicine,imaging,physicalprinciplesofrenewableenergyproduction,advancedprostheses, andenvironmentalcontrolandengineering. Editor-in-Chief: EliasGreenbaum,OakRidgeNationalLaboratory,OakRidge,Tennessee,USA EditorialBoard: JudithHerzfeld,DepartmentofChemistry, MasuoAizawa,DepartmentofBioengineering, BrandeisUniversity,Waltham,Massachusetts,USA TokyoInstituteofTechnology,Yokohama,Japan MarkS.Humayun,DohenyEyeInstitute, OlafS.Andersen,DepartmentofPhysiology, LosAngeles,California,USA Biophysics&MolecularMedicine, PierreJoliot,InstitutedeBiologiePhysico-Chimique, CornellUniversity,NewYork,USA FondationEdmonddeRothschild,Paris,France RobertH.Austin,DepartmentofPhysics, LajosKeszthelyi,InstituteofBiophysics, PrincetonUniversity,Princeton,NewJersey,USA HungarianAcademyofSciences,Szeged,Hungary JamesBarber,DepartmentofBiochemistry, RobertS.Knox,DepartmentofPhysics ImperialCollegeofScience,Technology andAstronomy,UniversityofRochester, andMedicine,London,England Rochester,NewYork,USA HowardC.Berg,DepartmentofMolecular AaronLewis,DepartmentofAppliedPhysics, andCellularBiology,HarvardUniversity, HebrewUniversity,Jerusalem,Israel Cambridge,Massachusetts,USA StuartM.Lindsay,DepartmentofPhysics VictorBloomfield,DepartmentofBiochemistry, andAstronomy,ArizonaStateUniversity, UniversityofMinnesota,St.Paul, Tempe,Arizona,USA Minnesota,USA DavidMauzerall,RockefellerUniversity, RobertCallender,DepartmentofBiochemistry, NewYork,NewYork,USA AlbertEinsteinCollegeofMedicine, EugenieV.Mielczarek,DepartmentofPhysics Bronx,NewYork,USA andAstronomy,GeorgeMasonUniversity, BrittonChance,DepartmentofBiochemistry/ Fairfax,Virginia,USA Biophysics,UniversityofPennsylvania, MarkolfNiemz,MedicalFacultyMannheim, Philadelphia,Pennsylvania,USA UniversityofHeidelberg,Mannheim,Germany StevenChu,LawrenceBerkeleyNational V.AdrianParsegian,PhysicalScienceLaboratory, Laboratory,Berkeley,California,USA NationalInstitutesofHealth,Bethesda, LouisJ.DeFelice,DepartmentofPharmacology, Maryland,USA VanderbiltUniversity,Nashville,Tennessee,USA LindaS.Powers,UniversityofArizona, JohannDeisenhofer,HowardHughesMedical Tucson,Arizona,USA Institute,TheUniversityofTexas,Dallas,Texas,USA EarlW.Prohofsky,DepartmentofPhysics, GeorgeFeher,DepartmentofPhysics,Universityof PurdueUniversity,WestLafayette,Indiana,USA California,SanDiego,LaJolla,California,USA AndrewRubin,DepartmentofBiophysics, HansFrauenfelder,LosAlamosNationalLaboratory, MoscowStateUniversity,Moscow,Russia LosAlamos,NewMexico,USA MichaelSeibert,NationalRenewableEnergy IvarGiaever,RensselaerPolytechnicInstitute, Laboratory,Golden,Colorado,USA Troy,NewYork,USA DavidThomas,DepartmentofBiochemistry, SolM.Gruner,CornellUniversity,Ithaca, UniversityofMinnesotaMedicalSchool, NewYork,USA Minneapolis,Minnesota,USA Forothertitlespublishedinthisseries,goto http://www.springer.com/series/3740 Geoff Dougherty Editor Medical Image Processing Techniques and Applications 123 Editor GeoffDougherty AppliedPhysicsandMedicalImaging CaliforniaStateUniversityChannelIslands OneUniversityDrive 93012Camarillo USA [email protected] ISSN1618-7210 ISBN978-1-4419-9769-2 e-ISBN978-1-4419-9779-1 DOI10.1007/978-1-4419-9779-1 SpringerNewYorkDordrechtHeidelbergLondon LibraryofCongressControlNumber:2011931857 (cid:2)c SpringerScience+BusinessMedia,LLC2011 Allrightsreserved.Thisworkmaynotbetranslatedorcopiedinwholeorinpartwithoutthewritten permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY10013, USA),except forbrief excerpts inconnection with reviews orscholarly analysis. Usein connectionwithanyformofinformationstorageandretrieval,electronicadaptation,computersoftware, orbysimilarordissimilarmethodologynowknownorhereafterdevelopedisforbidden. Theuseinthispublicationoftradenames,trademarks,servicemarks,andsimilarterms,eveniftheyare notidentifiedassuch,isnottobetakenasanexpressionofopinionastowhetherornottheyaresubject toproprietaryrights. Printedonacid-freepaper SpringerispartofSpringerScience+BusinessMedia(www.springer.com) To mymother,AdelineMaudDougherty, andmyfather,HarryDougherty(who left uson 17thNovember 2009) Preface The field of medical imaging advances so rapidly that all of those working in it, scientists, engineers, physicians, educators, and others, need to frequently update their knowledge to stay abreast of developments. While journals and periodicals play a crucial role in this, more extensive, integrative publications that connect fundamentalprinciplesandnewadvancesinalgorithmsandtechniquestopractical applicationsareessential.Suchpublicationshaveanextendedlifeandformdurable links in connecting past procedures to the present, and present procedures to the future.Thisbookaimstomeetthischallengeandprovideanenduringbridgeinthe everexpandingfieldofmedicalimaging. This book is designed for end users in the field of medical imaging, who wish to update their skills and understanding with the latest techniques in image analysis.Thebookemphasizestheconceptualframeworkofimageanalysisandthe effective use of image processing tools. It is designed to assist cross-disciplinary dialog both at graduate and at specialist levels, between all those involved in the multidisciplinary area of digital image processing, with a bias toward medical applications.Itsaimistoenablenewenduserstodrawontheexpertiseofexperts acrossthespecializationgap. To accomplish this, the book uses applicationsin a variety of fields to demon- strate and consolidate both specific and general concepts, and to build intuition, insight,andunderstanding.Itpresentsadetailedapproachtoeachapplicationwhile emphasizing the applicability of techniques to other research areas. Although the chapters are essentially self-contained, they reference other chapters to form an integrated whole. Each chapter uses a pedagogicalapproachto ensure conceptual learningbeforeintroducingspecifictechniquesand“tricksofthetrade”. The book is unified by the theme foreshadowed in the title “Medical Image Processing:TechniquesandApplications.”Itconsistsofacollectionofspecialized topics, each presented by a specialist in the field. Each chapter is split into sectionsand subsections, and beginswith an introductionto the topic,method,or technology.Emphasisisplacednotonlyonthebackgroundtheorybutalso onthe practicalaspects of the method, the details necessary to implementthe technique, vii viii Preface and limits of applicability. The chapter then introduces selected more advanced applicationsofthetopic,method,ortechnology,leadingtowardrecentachievements and unresolvedquestionsin a mannerthat can be understoodby a reader with no specialistknowledgeinthatarea. Chapter1,byDougherty,presentsabriefoverviewofmedicalimageprocessing. He outlinesa numberofchallengesandhighlightsopportunitiesforfurtherdevel- opment. A number of image analysis packages exist, both commercial and free, which makeuse oflibrariesof routinesthatcan be assembled/mobilized/concatenatedto automate an image analysis task. Chapter 2, by Luengo, Malm, and Bengtsson, introduces one such package, DIPimage, which is a toolbox for MatLab that incorporatesaGUIforautomaticimagedisplayandaconvenientdrop-downmenu of common image analysis functions. The chapter demonstrates how one can quicklydevelopasolutiontoautomateacommonassessmenttasksuchascounting cancerouscellsinaPapsmear. Segmentationisoneofthekeytoolsinmedicalimageanalysis.Themainappli- cationofsegmentationis in delineatinganorganreliably,quickly,andeffectively. Chapter 3, by Couprie, Najman and Talbot, presents very recent approaches that unifypopulardiscretesegmentationmethods. Deformable models are a promising method to handle the difficulties in seg- menting images that are contaminated by noise and sampling artifact. The model is represented by an initial curve (or surface in three dimensions (3D)) in the imagewhichevolvesundertheinfluenceofinternalenergy,derivedfromthemodel geometry, and an external force, defined from the image data. Segmentation is thenachievedbyminimizingthe sumofthese energies,whichusuallyresultsin a smoothcontour.InChapter4,Alfiansyahpresentsareviewofdifferentdeformable modelsandissuesrelatedtotheirimplementations.Hepresentssomeexamplesof thedifferentmodelsusedwithnoisymedicalimages. Over the past two decades, many authorshave investigatedthe use of MRI for the analysis of body fat and its distribution. However, when performedmanually, accurate isolation of fat in MR images can be an arduous task. In order to alleviate this burden,numeroussegmentation algorithmshave been developedfor the quantificationof fatin MR images.These includea numberofautomatedand semi-automatedsegmentationalgorithms.InChapter5,CostelloandKennydiscuss some of the techniques and models used in these algorithms, with a particular emphasisontheirapplicationandimplementation.Thepotentialimpactofartifacts suchasintensityinhomogeneities,partialvolumeeffect(PVE),andchemicalshift artifactsonimagesegmentationarealsodiscussed. An increasing portion of medical imaging problems concern thin objects, and particularlyvesselfiltering,segmentation,andclassification.Exampleapplications include vascular tree analysis in the brain, the heart, or the liver, the detection of aneurysms, stenoses, and arteriovenous malformations in the brain, and coronal treeanalysisinrelationtothepreventionofmyocardialinfarction.Thin,vessel-like objectsaremoredifficulttoprocessingeneralthanmostimagesfeatures,precisely becausetheyarethin.Theyarepronetodisappearwhenusingmanycommonimage Preface ix analysis operators, particularly in 3D. Chapter 6, by Tankyevych, Talbot, Passat, Musacchio, and Lagneau, introduces the problem of cerebral vessel filtering and detectionin3Danddescribesthestateoftheartfromfilteringtosegmentation,using localorientation,enhancement,localtopology,andscaleselection.Theyapplyboth linearandnonlinearoperatorstoatlascreation. Automated detection of linear structures is a common challenge in many computervisionapplications.Wheresuchstructuresoccurinmedicalimages,their measurement and interpretation are important steps in certain clinical decision- making tasks. In Chapter 7, Dabbah, Graham, Malik, and Efron discuss some of the well-knownlinear structure detectionmethodsused in medicalimaging.They describeaquantitativemethodforevaluatingtheperformanceofthesealgorithmsin comparisonwiththeirnewlydevelopedmethodfordetectingnervefibersinimages obtainedusinginvivocornealconfocalmicroscopy(CCM). Advances in linear feature detection have enabled new applications where the reliabletracingofline-likestructuresiscritical.Thisincludesneuriteidentification inimagesofbraincells,thecharacterizationofbloodvessels,thedelineationofcell membranes,andthesegmentationofbacteriaunderhighresolutionphasecontrast microscopy. Linear features represent fundamental image analysis primitives. In Chapter 8, Domanski, Sun, Lagerstrom, Wang, Bischof, Payne, and Vallotton introducethealgorithmsforlinearfeaturedetection,considerthepreprocessingand speed options, and show how such processing can be implemented conveniently usingagraphicaluserinterfacecalledHCA-Vision.Thechapterdemonstrateshow thirdpartiescanexploitthesenewcapabilitiesasinformedusers. Osteoporosisisadegenerativediseaseofthebone.Theaveragingnatureofbone mineral density measurement does not take into account the microarchitectural deterioration within the bone. In Chapter 9, Haidekker and Dougherty consider methodsthatallowthedegreeofmicroarchitecturaldeteriorationoftrabecularbone to be quantified.These havethe potentialto predictthe load-bearingcapabilityof bone. InChapter10,AdamandDoughertydescribetheapplicationofmedicalimage processingtotheassessmentandtreatmentofspinaldeformity,withafocusonthe surgical treatment of idiopathic scoliosis. The natural history of spinal deformity andcurrentapproachestosurgicalandnonsurgicaltreatmentarebrieflydescribed, followed by an overview of current clinically used imaging modalities. The key metricscurrentlyused to assess the severity and progressionof spinaldeformities from medical images are presented, followed by a discussion of the errors and uncertainties involved in manual measurements. This provides the context for an analysisofautomatedandsemi-automatedimageprocessingapproachestomeasure spinalcurveshapeandseverityintwoandthreedimensions. In Chapter 11, Cree and Jelinek outline the methods for acquiring and pre- processing of retinal images. They show how morphological,wavelet, and fractal methodscanbeusedtodetectlesionsandindicatethefuturedirectionsofresearch inthisarea. Theappearanceoftheretinalbloodvesselsisanimportantdiagnosticindicator for much systemic pathology. In Chapter 12, Iorga and Dougherty show that the
The list of books you might like
