2022
Mitigating bias in deep learning for diagnosis of coronary artery disease from myocardial perfusion SPECT images
Miller RJH, Singh A, Otaki Y, Tamarappoo BK, Kavanagh P, Parekh T, Hu LH, Gransar H, Sharir T, Einstein AJ, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli MF, Liang JX, Dey D, Berman DS, Slomka PJ. Mitigating bias in deep learning for diagnosis of coronary artery disease from myocardial perfusion SPECT images. European Journal Of Nuclear Medicine And Molecular Imaging 2022, 50: 387-397. PMID: 36194270, PMCID: PMC10042590, DOI: 10.1007/s00259-022-05972-w.Peer-Reviewed Original ResearchConceptsCoronary artery diseaseMyocardial perfusion imagingArtery diseaseInvasive angiographyObstructive coronary artery diseaseDisease probabilityLow-risk patientsLow-risk populationHigh-risk populationTotal perfusion deficitHigh diagnostic accuracyS-TPDPerfusion deficitsPatient managementPatientsPerfusion imagingDiagnostic accuracyPerfusion SPECT imagesLower likelihoodGood calibrationCharacteristic curveAngiographySPECT imagesSelection biasDisease
2021
Diagnostic safety of a machine learning-based automatic patient selection algorithm for stress-only myocardial perfusion SPECT
Eisenberg E, Miller RJH, Hu LH, Rios R, Betancur J, Azadani P, Han D, Sharir T, Einstein AJ, Bokhari S, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli M, Liang JX, Otaki Y, Tamarappoo BK, Dey D, Berman DS, Slomka PJ. Diagnostic safety of a machine learning-based automatic patient selection algorithm for stress-only myocardial perfusion SPECT. Journal Of Nuclear Cardiology 2021, 29: 2295-2307. PMID: 34228341, PMCID: PMC9020793, DOI: 10.1007/s12350-021-02698-4.Peer-Reviewed Original ResearchConceptsObstructive coronary artery diseaseCoronary artery diseaseHigh-risk coronary artery diseaseMyocardial perfusion imagingPatient selection algorithmTriple-vessel coronary artery diseasePrediction of CADML thresholdsStress-first protocolInvasive coronary angiographyReceiver operator characteristic curveOperator characteristic curveMyocardial perfusion SPECTArtery diseaseCoronary angiographyAnterior descendingClinical variablesClinical algorithmReader diagnosisRest imagingPerfusion SPECTPerfusion imagingDiagnostic safetyRadiation doseCharacteristic curve
2019
Machine learning predicts per-vessel early coronary revascularization after fast myocardial perfusion SPECT: results from multicentre REFINE SPECT registry.
Hu LH, Betancur J, Sharir T, Einstein AJ, Bokhari S, Fish MB, Ruddy TD, Kaufmann PA, Sinusas AJ, Miller EJ, Bateman TM, Dorbala S, Di Carli M, Germano G, Commandeur F, Liang JX, Otaki Y, Tamarappoo BK, Dey D, Berman DS, Slomka PJ. Machine learning predicts per-vessel early coronary revascularization after fast myocardial perfusion SPECT: results from multicentre REFINE SPECT registry. European Heart Journal - Cardiovascular Imaging 2019, 21: 549-559. PMID: 31317178, PMCID: PMC7167744, DOI: 10.1093/ehjci/jez177.Peer-Reviewed Original ResearchConceptsEarly coronary revascularizationMyocardial perfusion imagingStress TPDCoronary revascularizationTomography myocardial perfusion imagingTetrofosmin myocardial perfusion imagingSPECT myocardial perfusion imagingInvasive coronary angiographyReceiver operator characteristic curveREFINE SPECT registryPatient-specific explanationsOperator characteristic curveSingle photon emissionMyocardial perfusion SPECTCoronary angiographyIndividual patientsImaging variablesPatientsPerfusion SPECTPerfusion imagingClinical settingCharacteristic curveRevascularizationExpert interpretationStress test