Machine learning of ECG waveforms and cardiac magnetic resonance for response and survival after cardiac resynchronization therapy

• Published in: Computers in biology and medicine Vol. 178; p. 108627
• Main Authors: Bivona, Derek J., Ghadimi, Sona, Wang, Yu, Oomen, Pim J.A., Malhotra, Rohit, Darby, Andrew, Mangrum, J. Michael, Mason, Pamela K., Mazimba, Sula, Patel, Amit R., Epstein, Frederick H., Bilchick, Kenneth C.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.08.2024; Elsevier Limited
• Subjects: Accessibility; Cardiac resynchronization therapy; Congestive heart failure; Cost benefit analysis; EKG; Electrocardiogram; Electrocardiography; Functional principal component decomposition; Hazard mitigation; Heart failure; Human performance; Impact analysis; Learning algorithms; Machine learning; Magnetic resonance; Magnetic resonance imaging; Patients; Probabilistic models; Regression models; Survival; Waveforms; Heart failure; Functional principal component decomposition; Magnetic resonance imaging; Cardiac resynchronization therapy; Electrocardiogram; Machine learning
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2024.108627

Cardiac resynchronization therapy (CRT) can lead to marked symptom reduction and improved survival in selected patients with heart failure with reduced ejection fraction (HFrEF); however, many candidates for CRT based on clinical guidelines do not have a favorable response. A better way to identify patients expected to benefit from CRT that applies machine learning to accessible and cost-effective diagnostic tools such as the 12-lead electrocardiogram (ECG) could have a major impact on clinical care in HFrEF by helping providers personalize treatment strategies and avoid delays in initiation of other potentially beneficial treatments. This study addresses this need by demonstrating that a novel approach to ECG waveform analysis using functional principal component decomposition (FPCD) performs better than measures that require manual ECG analysis with the human eye and also at least as well as a previously validated but more expensive approach based on cardiac magnetic resonance (CMR). Analyses are based on five-fold cross validation of areas under the curve (AUCs) for CRT response and survival time after the CRT implant using Cox proportional hazards regression with stratification of groups using a Gaussian mixture model approach. Furthermore, FPCD and CMR predictors are shown to be independent, which demonstrates that the FPCD electrical findings and the CMR mechanical findings together provide a synergistic model for response and survival after CRT. In summary, this study provides a highly effective approach to prognostication after CRT in HFrEF using an accessible and inexpensive diagnostic test with a major expected impact on personalization of therapies. [Display omitted] •The 12-lead ECG is an inexpensive test that contains unrecognized information.•Machine learning can extract important prognostic information from ECG signals.•Functional principal component decomposition is effective for ECG waveform analysis.•ECG and MRI predictors of prognosis after CRT were independent.•The methods could be implemented in real-time during CRT procedures.