Physicians and Machine-Learning Algorithm Performance in Predicting Left-Ventricular Systolic Dysfunction from a Standard 12-Lead-Electrocardiogram.

Autor: Golany T; Taub Faculty of Computer Sciences, Technion-Israel Institute of Technology, Haifa 3200003, Israel., Radinsky K; Taub Faculty of Computer Sciences, Technion-Israel Institute of Technology, Haifa 3200003, Israel., Kofman N; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Litovchik I; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Young R; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Monayer A; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Love I; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Tziporin F; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Minha I; Department of Mathematics and Computer Science, The Open University, Raanana 4353701, Israel., Yehuda Y; Taub Faculty of Computer Sciences, Technion-Israel Institute of Technology, Haifa 3200003, Israel., Ziv-Baran T; Department of Epidemiology and Preventative Medicine, School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv 6997801, Israel., Fuchs S; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel., Minha S; Sackler School of Medicine, Tel-Aviv University, Ramat-Aviv, Tel Aviv 6997801, Israel.; Department of Cardiology, Shamir Medical Center, Be'er-Yaakov 7033001, Israel.
Jazyk: angličtina
Zdroj: Journal of clinical medicine [J Clin Med] 2022 Nov 15; Vol. 11 (22). Date of Electronic Publication: 2022 Nov 15.
DOI: 10.3390/jcm11226767
Abstrakt: Early detection of left ventricular systolic dysfunction (LVSD) may prompt early care and improve outcomes for asymptomatic patients. Standard 12-lead ECG may be used to predict LVSD. We aimed to compare the performance of Machine Learning Algorithms (MLA) and physicians in predicting LVSD from a standard 12-lead ECG. By utilizing a dataset of 13,820 pairs of ECGs and echocardiography, a deep residual convolutional neural network was trained for predicting LVSD (ejection fraction (EF) < 50%) from ECG. The ECGs of the test set (n = 850) were assessed for LVSD by the MLA and six physicians. The performance was compared using sensitivity, specificity, and C-statistics. The interobserver agreement between the physicians for the prediction of LVSD was moderate (κ = 0.50), with average sensitivity and specificity of 70%. The C-statistic of the MLA was 0.85. Repeating this analysis with LVSD defined as EF < 35% resulted in an improvement in physicians’ average sensitivity to 84% but their specificity decreased to 57%. The MLA C-statistic was 0.88 with this threshold. We conclude that although MLA outperformed physicians in predicting LVSD from standard ECG, prior to robust implementation of MLA in ECG machines, physicians should be encouraged to use this approach as a simple and readily available aid for LVSD screening.
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje