(cid:2) HEALTHCARE ANALYTICS (cid:2) (cid:2) (cid:2) (cid:2) (cid:2) (cid:2) Wiley Series in Operations Research and Management Science A complete list of the titles in this series appears at the end of this volume. (cid:2) (cid:2) HEALTHCARE ANALYTICS From Data to Knowledge to Healthcare Improvement HUIYANG Florida,USA EVAK.LEE (cid:2) Atlanta,USA (cid:2) (cid:2) (cid:2) Copyright©2016byJohnWiley&Sons,Inc.Allrightsreserved PublishedbyJohnWiley&Sons,Inc.,Hoboken,NewJersey PublishedsimultaneouslyinCanada Nopartofthispublicationmaybereproduced,storedinaretrievalsystem,ortransmittedinanyformor byanymeans,electronic,mechanical,photocopying,recording,scanning,orotherwise,exceptas permittedunderSection107or108ofthe1976UnitedStatesCopyrightAct,withouteithertheprior writtenpermissionofthePublisher,orauthorizationthroughpaymentoftheappropriateper-copyfeeto theCopyrightClearanceCenter,Inc.,222RosewoodDrive,Danvers,MA01923,(978)750-8400,fax (978)750-4470,oronthewebatwww.copyright.com.RequeststothePublisherforpermissionshould beaddressedtothePermissionsDepartment,JohnWiley&Sons,Inc.,111RiverStreet,Hoboken,NJ 07030,(201)748-6011,fax(201)748-6008,oronlineathttp://www.wiley.com/go/permission. LimitofLiability/DisclaimerofWarranty:Whilethepublisherandauthorhaveusedtheirbesteffortsin preparingthisbook,theymakenorepresentationsorwarrantieswithrespecttotheaccuracyor completenessofthecontentsofthisbookandspecificallydisclaimanyimpliedwarrantiesof merchantabilityorfitnessforaparticularpurpose.Nowarrantymaybecreatedorextendedbysales representativesorwrittensalesmaterials.Theadviceandstrategiescontainedhereinmaynotbesuitable foryoursituation.Youshouldconsultwithaprofessionalwhereappropriate.Neitherthepublishernor (cid:2) (cid:2) authorshallbeliableforanylossofprofitoranyothercommercialdamages,includingbutnotlimitedto special,incidental,consequential,orotherdamages. Forgeneralinformationonourotherproductsandservicesorfortechnicalsupport,pleasecontactour CustomerCareDepartmentwithintheUnitedStatesat(800)762-2974,outsidetheUnitedStatesat (317)572-3993orfax(317)572-4002. Wileyalsopublishesitsbooksinavarietyofelectronicformats.Somecontentthatappearsinprintmay notbeavailableinelectronicformats.FormoreinformationaboutWileyproducts,visitourwebsiteat www.wiley.com. LibraryofCongressCataloging-in-PublicationData: Names:Yang,Hui,1981-author.|Lee,EvaK.,author. Title:Healthcareanalytics:fromdatatoknowledgetohealthcare improvement/HuiYang,EvaK.Lee. Description:Hoboken,NewJersey:JohnWiley&Sons,2016.|Includes bibliographicalreferencesandindex. Identifiers:LCCN2015047966|ISBN9781118919392(cloth)|ISBN9781118919408 (online)|ISBN9781119374664(ePDF)|ISBN9781119374640(ePub) Subjects:LCSH:Medicalcare–Dataprocessing.|Medicalcare–Information services. Classification:LCCR858.A1Y362016|DDC362.10285–dc23LCrecordavailableat http://lccn.loc.gov/2015047966 Typesetin10/12ptTimesLTStdbySPiGlobal,Chennai,India PrintedintheUnitedStatesofAmerica (cid:2) (cid:2) CONTENTS LISTOFCONTRIBUTORS xvii PREFACE xxi (cid:2) (cid:2) PARTI ADVANCESINBIOMEDICALANDHEALTH INFORMATICS 1 1 RecentDevelopmentinMethodologyforGeneNetworkProblems andInferences 3 SungW.HanandHuaZhong 1.1 Introduction 3 1.2 Background 5 1.3 GeneticDataAvailable 7 1.4 Methodology 7 1.4.1 StructuralEquationModel 8 1.4.2 ScoreFunctionFormulation 9 1.4.3 Two-StageLearning 12 1.4.4 FurtherIssues 13 1.5 SearchAlgorithm 13 1.5.1 GlobalOptimalSolutionSearch 13 1.5.2 HeuristicAlgorithmforaLocalOptimalSolution 14 1.6 PCAlgorithm 15 (cid:2) (cid:2) vi CONTENTS 1.7 Application/CaseStudies 16 1.7.1 SkinCutaneousMelanoma(SKCM)Datafromthe TCGADataPortalWebsite 16 1.7.2 TheCCLE(CancerCellLineEncyclopedia)Project 20 1.7.3 CellularSignalingNetworkinFlowCytometryData 20 1.8 Discussion 23 1.9 OtherUsefulSoftwares 23 Acknowledgments 24 References 24 2 BiomedicalAnalyticsandMorphoproteomics:AnIntegrative ApproachforMedicalDecisionMakingforRecurrentor RefractoryCancers 31 MaryF.McGuireandRobertE.Brown 2.1 Introduction 31 2.2 Background 32 2.2.1 Data 33 2.2.2 Tools 33 2.2.3 Algorithms 34 2.2.4 LiteratureReview 35 2.3 Methodology 37 (cid:2) (cid:2) 2.3.1 Morphoproteomics(Fig.2.1(1–3)) 39 2.3.2 BiomedicalAnalytics(Fig.2.1(4–10)) 40 2.3.3 IntegratingMorphoproteomicsandBiomedical Analytics 44 2.4 CaseStudies 46 2.4.1 Clinical:TherapeuticRecommendationsforPancreatic Adenocarcinoma 46 2.4.2 Clinical:BiologyUnderlyingExceptionalResponder inRefractoryHodgkin’sLymphoma 48 2.4.3 Research:RoleoftheHypoxiaPathwayinBoth OncogenesisandEmbryogenesis 50 2.5 Discussion 51 2.6 Conclusions 52 Acknowledgments 53 References 53 3 CharacterizationandMonitoringofNonlinearDynamicsandChaos inComplexPhysiologicalSystems 59 HuiYang,YunChen,andFabioLeonelli 3.1 Introduction 59 3.2 Background 61 3.3 Sensor-BasedCharacterizationandModelingofNonlinear Dynamics 65 (cid:2) (cid:2) CONTENTS vii 3.3.1 MultifractalSpectrumAnalysisofNonlinearTime Series 65 3.3.2 RecurrenceQuantificationAnalysis 75 3.3.3 MultiscaleRecurrenceQuantificationAnalysis 78 3.4 HealthcareApplications 80 3.4.1 NonlinearCharacterizationofHeartRateVariability 81 3.4.2 MultiscaleRecurrenceAnalysisofSpace–Time PhysiologicalSignals 85 3.5 Summary 88 Acknowledgments 90 References 90 4 StatisticalModelingofElectrocardiographySignalforSubject MonitoringandDiagnosis 95 LiliChen,ChangyueSong,andXiZhang 4.1 Introduction 95 4.2 BasicElementsofECG 96 4.3 StatisticalModelingofECGforDiseaseDiagnosis 99 4.3.1 ECGSignalDenoising 100 4.3.2 WaveformDetection 105 4.3.3 FeatureExtraction 106 (cid:2) (cid:2) 4.3.4 DiseaseClassificationandDiagnosis 111 4.4 AnExample:DetectionofObstructiveSleepApneafrom aSingleECGLead 115 4.4.1 IntroductiontoObstructiveSleepApnea 115 4.5 MaterialsandMethods 115 4.5.1 Database 115 4.5.2 QRSDetectionandRRCorrection 116 4.5.3 RWaveAmplitudesandEDRSignal 117 4.5.4 FeatureSet 117 4.5.5 ClassifierTrainingwithFeatureSelection 118 4.6 Results 118 4.6.1 QRSDetectionandRRCorrection 118 4.6.2 FeatureSelection 118 4.6.3 OSADetection 120 4.7 ConclusionsandDiscussions 121 References 121 5 ModelingandSimulationofMeasurementUncertaintyinClinical Laboratories 127 VarunRamamohan,JamesT.Abbott,andYuehwernYih 5.1 Introduction 127 5.2 BackgroundandLiteratureReview 129 (cid:2) (cid:2) viii CONTENTS 5.2.1 MeasurementUncertainty:BackgroundandAnalytical Estimation 130 5.2.2 UncertaintyinClinicalLaboratories 134 5.2.3 UncertaintyinClinicalLaboratories:ASystem Approach 136 5.3 ModelDevelopmentGuidelines 138 5.3.1 SystemDescriptionandProcessPhases 138 5.3.2 ModelingGuidelines 139 5.4 ImplementationofGuidelines:EnzymeAssayUncertainty Model 141 5.4.1 CalibrationPhase 142 5.4.2 SampleAnalysisPhase 149 5.4.3 ResultsandAnalysis 150 5.5 DiscussionandConclusions 152 References 154 6 PredictiveAnalytics:ClassificationinMedicineandBiology 159 EvaK.Lee 6.1 Introduction 159 6.2 Background 161 6.3 MachineLearningwithDiscreteSupportVectorMachine (cid:2) PredictiveModels 163 (cid:2) 6.3.1 ModelingofReserved-JudgmentRegionfor GeneralGroups 164 6.3.2 DiscriminantAnalysisviaMixed-IntegerProgramming 165 6.3.3 ModelVariations 167 6.3.4 TheoreticalPropertiesandComputationalStrategies 170 6.4 ApplyingDAMIPtoReal-WorldApplications 170 6.4.1 ValidationofModelandComputationalEffort 171 6.4.2 ApplicationstoBiologicalandMedicalProblems 171 6.5 SummaryandConclusion 182 Acknowledgments 183 References 183 7 PredictiveModelinginRadiationOncology 189 HaoZhang,RobertMeyer,LeyuanShi,WeiLu,andWarrenD’Souza 7.1 Introduction 189 7.2 TutorialsofPredictiveModelingTechniques 191 7.2.1 FeatureSelection 191 7.2.2 SupportVectorMachine 192 7.2.3 LogisticRegression 193 7.2.4 DecisionTree 193 7.3 ReviewofRecentPredictiveModelingApplicationsinRadiation Oncology 194 7.3.1 MachineLearningforMedicalImageProcessing 194 (cid:2)