Bio-informatics assessment schema (BIAS) to improve myocardial perfusion image diagnostic and prognostic value: the NHLB...

MISSING IMAGE

Material Information

Title:
Bio-informatics assessment schema (BIAS) to improve myocardial perfusion image diagnostic and prognostic value: the NHLBI-sponsored women’s ischemia syndrome evaluation (WISE) study
Series Title:
Journal of Cardiovascular Magnetic Resonance
Physical Description:
Mixed Material
Creator:
Mark Doyle
Gerald Pohost
Leslee J Shaw
Diane V Thompson
Sheryl F Kelsey
B Delia Johnson
William J Rogers
Geetha Rayarao
Barry L Sharaf
Carl J Pepine
C Noel Bairey Merz
Robert W Biederman
Publisher:
Journal of Cardiovascular Magnetic Resonance
Publication Date:

Notes

General Note:
Poster Presentation

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
System ID:
AA00020085:00001

Full Text

PAGE 1

POSTERPRESENTATION OpenAccessBio-informaticsassessmentschema(BIAS)to improvemyocardialperfusionimagediagnosticand prognosticvalue:theNHLBI-sponsoredwomen s ischemiasyndromeevaluation(WISE)studyMarkDoyle1*,GeraldPohost7,LesleeJShaw8,DianeVThompson1,SherylFKelsey2,BDeliaJohnson2, WilliamJRogers4,GeethaRayarao1,BarryLSharaf5,CarlJPepine6,CNoelBaireyMerz3,RobertWBiederman1From 17thAnnualSCMRScientificSessions NewOrleans,LA,USA.16-19January2014BackgroundIntroduction:Whenassessingmyocardialperfusionimage (MPI)datafordiagnosisorpredictionofprognosisitis commontoevaluatethetechnologyusingaReceiverOperatorCharacteristic (ROC)curve.Thediagonal responserepresentstheknowledgepriortoconductingthe test(i.e.thediagonalrepresentsrandomchance).WepresentanapproachtermedBio-InformaticsAssessment Schema(BIAS)thatprovidesanelevatedbaselinefrom whichtoevaluatetheMPIreading,i.e.indicatingthatat baseline,moredataisavailablethanrandomchance.Here wedescribehowtheBIASformulaearegeneratedforcardiovascularmagneticresonanceimage(CMRI)dataapplied totheWomen sIschemiaSyndromeEvaluation(WISE) Study.MethodsMethods:Women(n=182),meanage5911yrs,with symptomssuggestiveofmyocar dialischemiaunderwent MPIandcardiacfunctionevaluationseparatelybyCMRI andSinglePositronComputedTomography(SPECT)and werefollowed(4017months)todocumentMACE(CV death,myocardialinfarction,andhospitalizationforcongestiveheartfailure).Abnormalperfusiondefectswere notedforeachMPImodalityusingclinicalcriteria(clinical reading)andifatleastoneregionwasabnormal,thestudy wasconsideredtobepositivefordisease.Multiplelinear regressionmodelsweregene rated,eachpredictingMPI statusfromonemodality(e.g.CMRI)usingdatafroma secondmodality(e.g.SPECT).TheBIASmodelswere generatedbyremovingthemy ocardialperfusionstatus fromthelinearregressionmodels,i.e.theBIASequations didnotcontainanyknowledgeoftheMPIstatus(e.g. BIASmodel=-.43+endsystolicvolumeindex0.011+ averagemyocardialwallthickness0.037).Thearea underthecurve(AUC)forROCanalysiswascalculated forthenaturalMPIstatusoftheCMRIdataandforthe BIAScurve.ROCanalysiswasconductedfordetectionof obstructivecoronaryarterydisease(CAD 50%stenosis) andpredictionofMACE.ResultsResults:CADwaspresentin57(31%)andMACE occurredin22women(12%).ForpredictionofCAD,the AUCswere:clinicalMPIreading0.68andBIASreading 0.67(Figure1).ForpredictionofMACE,theAUCswere: clinicalMPIreading0.54andBIASreading0.77(Figure2).ConclusionsConclusions:BIASmodeling,incorporatingcardiacvariablesmeasuredbyCMRI,producedanelevatedbaseline onROCanalysis(diagnosticandprognostic)thatwas higherthantheclinicalMPIreadinginprognosticcase. ThesignificanceofthisisthatthiselevatedbaselinerepresentstheinformationstatuspriortoreadingtheMPIdata. ThenextlogicalstepistoincorporatethisbaselineknowledgeintoreadingtheMPIstatus,whichisexpectedto resultindramaticimprovementofMPIdata. 1CardiacMRI,AlleghenyGeneralHospital,Pittsburgh,Pennsylvania,USA FulllistofauthorinformationisavailableattheendofthearticleDoyle etal JournalofCardiovascularMagnetic Resonance 2014, 16 (Suppl1):P201 http://www.jcmr-online.com/content/16/S1/P201 2014Doyleetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.TheCreativeCommonsPublicDomainDedicationwaiver(http:// creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,unlessotherwisestated.

PAGE 2

FundingThisworkwassupportedbycontractsfromtheNational Heart,LungandBloodInstitutesnos.N01-HV-68161, N01-HV-68162,N01-HV-68163,N01-HV-68164,grants U0164829,U01HL649141,U01HL649241,T32HL69751, R01HL090957,1R03AG032631fromtheNationalInstituteonAging,GCRCgrantMO1-RR00425fromthe NationalCenterforResearchResources,theNational CenterforAdvancingTranslationalSciencesGrant UL1TR000124,andgrantsfromtheGustavusandLouis PfeifferResearchFoundation,Danville,NJ,TheWomen s GuildofCedars-SinaiMedicalCenter,LosAngeles,CA, TheLadiesHospitalAidSocietyofWesternPennsylvania, Pittsburgh,PA,andQMED,Inc.,LaurenceHarbor,NJ, theEdytheL.BroadWomen sHeartResearchFellowship, Cedars-SinaiMedicalCenter,LosAngeles,California,the BarbraStreisandWomen sCardiovascularResearchand EducationProgram,Cedars-SinaiMedicalCenter,Los AngelesandTheSocietyforWomen sHealthResearch (SWHR),Washington,D.C.Authors details1CardiacMRI,AlleghenyGeneralHospital,Pittsburgh,Pennsylvania,USA.2Cardiology,UniversityofPittsburgh,Pittsburgh,Pennsylvania,USA.3Cardiology,Cedars-SinaiMedicalCtr,LosAngeles,California,USA.4Cardiology,Univ.ofAlabama,Birmingham,Alabama,USA.5Cardiology, BrownUniversity,Providence,RhodeIsland,USA.6Cardiology,Universityof Florida,Gainesville,Florida,USA.7Cardiology,Univ.ofSouthernCalifornia,Los Angeles,California,USA.8Cardiology,AtlantaCardiovascularResearchInst., Atlanta,Georgia,USA. Published:16January2014doi:10.1186/1532-429X-16-S1-P201 Citethisarticleas: Doyle etal .: Bio-informaticsassessmentschema(BIAS) toimprovemyocardialperfusionimagediagnosticandprognosticvalue: theNHLBI-sponsoredwomen sischemiasyndromeevaluation(WISE) study. JournalofCardiovascularMagneticResonance 2014 16 (Suppl1):P201. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Figure 1 ROC of CAD Detection Figure 2 ROC of MACE Detection Doyle et al Journal of Cardiovascu lar Magnetic Resonance 2014, 16 (Suppl 1):P201 http://ww w.jcmr-online .com/conten t/16/S1/P20 1 Page 2 of 2