Machine Learned Cellular Phenotypes in Cardiomyopathy Predict Sudden Death
Standard
Machine Learned Cellular Phenotypes in Cardiomyopathy Predict Sudden Death. / Rogers, Albert J; Selvalingam, Anojan; Alhusseini, Mahmood I; Krummen, David E; Corrado, Cesare; Abuzaid, Firas; Baykaner, Tina; Meyer, Christian; Clopton, Paul; Giles, Wayne; Bailis, Peter; Niederer, Steven; Wang, Paul J; Rappel, Wouter-Jan; Zaharia, Matei; Narayan, Sanjiv M.
In: CIRC RES, Vol. 128, No. 2, 22.01.2021, p. 172-184.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Machine Learned Cellular Phenotypes in Cardiomyopathy Predict Sudden Death
AU - Rogers, Albert J
AU - Selvalingam, Anojan
AU - Alhusseini, Mahmood I
AU - Krummen, David E
AU - Corrado, Cesare
AU - Abuzaid, Firas
AU - Baykaner, Tina
AU - Meyer, Christian
AU - Clopton, Paul
AU - Giles, Wayne
AU - Bailis, Peter
AU - Niederer, Steven
AU - Wang, Paul J
AU - Rappel, Wouter-Jan
AU - Zaharia, Matei
AU - Narayan, Sanjiv M
PY - 2021/1/22
Y1 - 2021/1/22
N2 - RATIONALE: Susceptibility to VT/VF (ventricular tachycardia/fibrillation) is difficult to predict in patients with ischemic cardiomyopathy either by clinical tools or by attempting to translate cellular mechanisms to the bedside.OBJECTIVE: To develop computational phenotypes of patients with ischemic cardiomyopathy, by training then interpreting machine learning of ventricular monophasic action potentials (MAPs) to reveal phenotypes that predict long-term outcomes.METHODS AND RESULTS: We recorded 5706 ventricular MAPs in 42 patients with coronary artery disease and left ventricular ejection fraction ≤40% during steady-state pacing. Patients were randomly allocated to independent training and testing cohorts in a 70:30 ratio, repeated K=10-fold. Support vector machines and convolutional neural networks were trained to 2 end points: (1) sustained VT/VF or (2) mortality at 3 years. Support vector machines provided superior classification. For patient-level predictions, we computed personalized MAP scores as the proportion of MAP beats predicting each end point. Patient-level predictions in independent test cohorts yielded c-statistics of 0.90 for sustained VT/VF (95% CI, 0.76-1.00) and 0.91 for mortality (95% CI, 0.83-1.00) and were the most significant multivariate predictors. Interpreting trained support vector machine revealed MAP morphologies that, using in silico modeling, revealed higher L-type calcium current or sodium-calcium exchanger as predominant phenotypes for VT/VF.CONCLUSIONS: Machine learning of action potential recordings in patients revealed novel phenotypes for long-term outcomes in ischemic cardiomyopathy. Such computational phenotypes provide an approach which may reveal cellular mechanisms for clinical outcomes and could be applied to other conditions.
AB - RATIONALE: Susceptibility to VT/VF (ventricular tachycardia/fibrillation) is difficult to predict in patients with ischemic cardiomyopathy either by clinical tools or by attempting to translate cellular mechanisms to the bedside.OBJECTIVE: To develop computational phenotypes of patients with ischemic cardiomyopathy, by training then interpreting machine learning of ventricular monophasic action potentials (MAPs) to reveal phenotypes that predict long-term outcomes.METHODS AND RESULTS: We recorded 5706 ventricular MAPs in 42 patients with coronary artery disease and left ventricular ejection fraction ≤40% during steady-state pacing. Patients were randomly allocated to independent training and testing cohorts in a 70:30 ratio, repeated K=10-fold. Support vector machines and convolutional neural networks were trained to 2 end points: (1) sustained VT/VF or (2) mortality at 3 years. Support vector machines provided superior classification. For patient-level predictions, we computed personalized MAP scores as the proportion of MAP beats predicting each end point. Patient-level predictions in independent test cohorts yielded c-statistics of 0.90 for sustained VT/VF (95% CI, 0.76-1.00) and 0.91 for mortality (95% CI, 0.83-1.00) and were the most significant multivariate predictors. Interpreting trained support vector machine revealed MAP morphologies that, using in silico modeling, revealed higher L-type calcium current or sodium-calcium exchanger as predominant phenotypes for VT/VF.CONCLUSIONS: Machine learning of action potential recordings in patients revealed novel phenotypes for long-term outcomes in ischemic cardiomyopathy. Such computational phenotypes provide an approach which may reveal cellular mechanisms for clinical outcomes and could be applied to other conditions.
KW - Action Potentials
KW - Aged
KW - Aged, 80 and over
KW - Cardiomyopathies/diagnosis
KW - Death, Sudden, Cardiac/etiology
KW - Diagnosis, Computer-Assisted
KW - Electrophysiologic Techniques, Cardiac
KW - Female
KW - Humans
KW - Male
KW - Middle Aged
KW - Myocardial Infarction/complications
KW - Neural Networks, Computer
KW - Phenotype
KW - Predictive Value of Tests
KW - Prognosis
KW - Prospective Studies
KW - Risk Assessment
KW - Risk Factors
KW - Signal Processing, Computer-Assisted
KW - Support Vector Machine
KW - Tachycardia, Ventricular/diagnosis
KW - Time Factors
KW - Ventricular Fibrillation/diagnosis
U2 - 10.1161/CIRCRESAHA.120.317345
DO - 10.1161/CIRCRESAHA.120.317345
M3 - SCORING: Journal article
C2 - 33167779
VL - 128
SP - 172
EP - 184
JO - CIRC RES
JF - CIRC RES
SN - 0009-7330
IS - 2
ER -