2024
Deep learning-based epicardial adipose tissue measurement, maximizing prognostic information from attenuation correction imaging
Shanbhag A, Miller R, Killekar A, Lemley M, Bednarski B, Van Kriekinge S, Kavanagh P, Feher A, Miller E, Bateman T, Liang J, Builoff V, Berman D, Dey D, Slomka P. Deep learning-based epicardial adipose tissue measurement, maximizing prognostic information from attenuation correction imaging. Progress In Biomedical Optics And Imaging 2024, 12930: 129300b-129300b-6. DOI: 10.1117/12.3007914.Peer-Reviewed Original ResearchEpicardial adipose tissueUngated CTMyocardial infarctionEAT volumeRisk stratificationComputed tomographyEAT volume measurementEpicardial adipose tissue measurementsEpicardial adipose tissue volumeAssociated with risk of cardiovascular eventsRisk of cardiovascular eventsMedian follow-upIncreased risk of deathGated computed tomographyImprove risk stratificationAdipose tissue measurementsAssociated with riskRisk of deathAttenuation correction imagesGated CTChest CTNo significant differencePrognostic informationCardiovascular eventsFollow-upClinical phenotypes among patients with normal cardiac perfusion using unsupervised learning: a retrospective observational study
Miller R, Bednarski B, Pieszko K, Kwiecinski J, Williams M, Shanbhag A, Liang J, Huang C, Sharir T, Hauser M, Dorbala S, Di Carli M, Fish M, Ruddy T, Bateman T, Einstein A, Kaufmann P, Miller E, Sinusas A, Acampa W, Han D, Dey D, Berman D, Slomka P. Clinical phenotypes among patients with normal cardiac perfusion using unsupervised learning: a retrospective observational study. EBioMedicine 2024, 99: 104930. PMID: 38168587, PMCID: PMC10794922, DOI: 10.1016/j.ebiom.2023.104930.Peer-Reviewed Original ResearchMyocardial infarctionMyocardial perfusion imagingBlood InstituteNational HeartPharmacologic stressExternal testing cohortNormal imaging resultsRetrospective observational studyCoronary artery diseasePrevious myocardial infarctionRisk of deathNormal perfusion scanBritish Heart FoundationNational InstituteCluster 4 patientsDistinct phenotypesCardiovascular riskArtery diseaseRisk stratificationPerfusion scanNormal perfusionImaging featuresNormal scansMPI patientsHeart Foundation
2023
Unsupervised learning to characterize patients with known coronary artery disease undergoing myocardial perfusion imaging
Williams M, Bednarski B, Pieszko K, Miller R, Kwiecinski J, Shanbhag A, Liang J, Huang C, Sharir T, Dorbala S, Di Carli M, Einstein A, Sinusas A, Miller E, Bateman T, Fish M, Ruddy T, Acampa W, Hauser M, Kaufmann P, Dey D, Berman D, Slomka P. Unsupervised learning to characterize patients with known coronary artery disease undergoing myocardial perfusion imaging. European Journal Of Nuclear Medicine And Molecular Imaging 2023, 50: 2656-2668. PMID: 37067586, PMCID: PMC10317876, DOI: 10.1007/s00259-023-06218-z.Peer-Reviewed Original ResearchConceptsCoronary artery diseaseMyocardial perfusion imagingCause mortalityTotal perfusion deficitArtery diseaseRisk stratificationPerfusion deficitsHigher body mass indexTomography myocardial perfusion imagingMore diabetes mellitusSPECT myocardial perfusion imagingBetter risk stratificationRisk-stratify patientsBody mass indexREFINE SPECT registrySingle photon emissionDiabetes mellitusMass indexExternal cohortStress imagingMyocardial perfusionPatientsRisk phenotypePerfusion imagingImaging characteristics
2022
Direct Risk Assessment From Myocardial Perfusion Imaging Using Explainable Deep Learning
Singh A, Miller RJH, Otaki Y, Kavanagh P, Hauser MT, Tzolos E, Kwiecinski J, Van Kriekinge S, Wei CC, Sharir T, Einstein AJ, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli M, Liang JX, Huang C, Han D, Dey D, Berman DS, Slomka PJ. Direct Risk Assessment From Myocardial Perfusion Imaging Using Explainable Deep Learning. JACC Cardiovascular Imaging 2022, 16: 209-220. PMID: 36274041, PMCID: PMC10980287, DOI: 10.1016/j.jcmg.2022.07.017.Peer-Reviewed Original ResearchConceptsMyocardial perfusion imagingTotal perfusion deficitNonfatal myocardial infarctionMyocardial infarctionPerfusion imagingTomography myocardial perfusion imagingIschemic total perfusion deficitStress total perfusion deficitTesting groupReceiver-operating characteristic curvePatient-level riskPrediction of deathSingle photon emissionLogistic regression modelsCause mortalityPrimary outcomeHighest quartileRisk stratificationAbnormal perfusionNormal perfusionPerfusion deficitsAdverse event predictionPrognostic accuracyHigh riskMyocardial perfusionUnsupervised machine learning improves risk stratification of patients with visual normal SPECT myocardial perfusion imaging assessments
Bednarski B, Williams M, Pieszko K, Miller R, Huang C, Kwiecinski J, Sharir T, Di Carli M, Fish M, Ruddy T, Hasuer T, Miller E, Acampa W, Berman D, Slomka P. Unsupervised machine learning improves risk stratification of patients with visual normal SPECT myocardial perfusion imaging assessments. European Heart Journal 2022, 43: ehac544.300. DOI: 10.1093/eurheartj/ehac544.300.Peer-Reviewed Original ResearchMajor adverse cardiac eventsPeak systolic blood pressureSystolic blood pressureHigh-risk clustersRisk stratificationMyocardial perfusion imagingNormal perfusionHazard ratioBlood pressureCox proportional hazards analysisHigher left ventricular massHigher body mass indexTomography myocardial perfusion imagingNormal clinical assessmentRisk-stratified subgroupsAdverse cardiac eventsPrevalence of diabetesProportional hazards analysisBetter risk stratificationCoronary artery diseaseImproved risk stratificationManagement of patientsKaplan-Meier curvesBody mass indexLeft ventricular mass
2021
Determining a minimum set of variables for machine learning cardiovascular event prediction: results from REFINE SPECT registry
Rios R, Miller RJH, Hu LH, Otaki Y, Singh A, Diniz M, Sharir T, Einstein AJ, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, DiCarli M, Van Kriekinge S, Kavanagh P, Parekh T, Liang JX, Dey D, Berman DS, Slomka P. Determining a minimum set of variables for machine learning cardiovascular event prediction: results from REFINE SPECT registry. Cardiovascular Research 2021, 118: 2152-2164. PMID: 34259870, PMCID: PMC9302886, DOI: 10.1093/cvr/cvab236.Peer-Reviewed Original ResearchConceptsMajor adverse cardiac eventsMyocardial perfusion imagingTotal perfusion deficitPrognostic accuracyREFINE SPECT registryImaging variablesTraditional multivariable modelsMultivariable modelStress total perfusion deficitAdverse cardiac eventsReceiver-operating characteristic curveOptimal risk stratificationTomography (SPECT) MPIComparable prognostic accuracyComparable prognostic performanceHigher prognostic accuracySingle photon emissionCardiovascular event predictionCardiac eventsRisk stratificationClinical variablesPerfusion deficitsPrognostic performancePerfusion imagingCollected variables
2020
Automated quantitative analysis of CZT SPECT stratifies cardiovascular risk in the obese population: Analysis of the REFINE SPECT registry
Klein E, Miller RJH, Sharir T, Einstein AJ, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli M, Otaki Y, Gransar H, Liang JX, Dey D, Berman DS, Slomka PJ. Automated quantitative analysis of CZT SPECT stratifies cardiovascular risk in the obese population: Analysis of the REFINE SPECT registry. Journal Of Nuclear Cardiology 2020, 29: 727-736. PMID: 32929639, PMCID: PMC8497048, DOI: 10.1007/s12350-020-02334-7.Peer-Reviewed Original ResearchConceptsBody mass indexTotal perfusion deficitStress total perfusion deficitMyocardial perfusion imagingSPECT myocardial perfusion imagingPrognostic accuracyObese populationMajor adverse cardiac event risksCox proportional hazards analysisDifferent obesity categoriesRobust risk stratificationCardiac event riskProportional hazards analysisHigher prognostic accuracyREFINE SPECT registrySoft tissue attenuationQuantitative perfusion analysisMACE riskObesity categoriesCardiovascular riskObese patientsAdjusted analysisMass indexRisk stratificationSignificant obesity
2019
5-Year Prognostic Value of Quantitative Versus Visual MPI in Subtle Perfusion Defects Results From REFINE SPECT
Otaki Y, Betancur J, Sharir T, Hu LH, Gransar H, Liang JX, Azadani PN, Einstein AJ, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli M, Tamarappoo BK, Germano G, Dey D, Berman DS, Slomka PJ. 5-Year Prognostic Value of Quantitative Versus Visual MPI in Subtle Perfusion Defects Results From REFINE SPECT. JACC Cardiovascular Imaging 2019, 13: 774-785. PMID: 31202740, PMCID: PMC6899217, DOI: 10.1016/j.jcmg.2019.02.028.Peer-Reviewed Original ResearchConceptsMajor adverse cardiac eventsTotal perfusion deficitStress total perfusion deficitMACE rateHazard ratioRate of MACECox proportional hazards analysisTomography myocardial perfusion imagingAdjusted hazard ratioAdverse cardiac eventsNonfatal myocardial infarctionProportional hazards analysisMyocardial perfusion imagingSingle photon emissionCardiac eventsLate revascularizationUnstable anginaRisk stratificationPrognostic valueKaplan-MeierPerfusion abnormalitiesPerfusion deficitsMyocardial infarctionPrognostic studiesMyocardial perfusion
2016
Visual identification of coronary calcifications on attenuation correction CT improves diagnostic accuracy of SPECT/CT myocardial perfusion imaging
Patchett ND, Pawar S, Miller EJ. Visual identification of coronary calcifications on attenuation correction CT improves diagnostic accuracy of SPECT/CT myocardial perfusion imaging. Journal Of Nuclear Cardiology 2016, 24: 711-720. PMID: 26850031, DOI: 10.1007/s12350-016-0395-5.Peer-Reviewed Original ResearchMeSH KeywordsArtifactsComputed Tomography AngiographyCoronary Artery DiseaseFemaleHumansImage EnhancementMaleMiddle AgedMyocardial Perfusion ImagingObserver VariationReproducibility of ResultsSensitivity and SpecificitySingle Photon Emission Computed Tomography Computed TomographyUser-Computer InterfaceVascular CalcificationConceptsCoronary artery calciumMyocardial perfusion imagingCT myocardial perfusion imagingSPECT/CT myocardial perfusion imagingInvasive coronary angiographyAttenuation correction CTCause mortalityLate revascularizationAbsence of CACAccuracy of MPIRetrospective chart reviewFalse-positive studiesCT myocardial perfusionArtery calciumChart reviewHazard ratioConsecutive patientsCoronary angiographyCoronary calcificationRisk stratificationRisk markersMore MIsCT scanMyocardial perfusionPatients