Detecting cardiomyopathies in pregnancy and the postpartum period with an electrocardiogram-based deep learning model

Abstract

Aims: Cardiovascular disease is a major threat to maternal health, with cardiomyopathy being among the most common acquired cardiovascular diseases during pregnancy and the postpartum period. The aim of our study was to evaluate the effectiveness of an electrocardiogram (ECG)-based deep learning model in identifying cardiomyopathy during pregnancy and the postpartum period.

Methods and results: We used an ECG-based deep learning model to detect cardiomyopathy in a cohort of women who were pregnant or in the postpartum period seen at Mayo Clinic. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. We compared the diagnostic probabilities of the deep learning model with natriuretic peptides and a multivariable model consisting of demographic and clinical parameters. The study cohort included 1807 women; 7%, 10%, and 13% had left ventricular ejection fraction (LVEF) of 35% or less, <45%, and <50%, respectively. The ECG-based deep learning model identified cardiomyopathy with AUCs of 0.92 (LVEF ≤ 35%), 0.89 (LVEF < 45%), and 0.87 (LVEF < 50%). For LVEF of 35% or less, AUC was higher in Black (0.95) and Hispanic (0.98) women compared to White (0.91). Natriuretic peptides and the multivariable model had AUCs of 0.85 to 0.86 and 0.72, respectively.

Conclusions: An ECG-based deep learning model effectively identifies cardiomyopathy during pregnancy and the postpartum period and outperforms natriuretic peptides and traditional clinical parameters with the potential to become a powerful initial screening tool for cardiomyopathy in the obstetric care setting.

Keywords: Artificial intelligence; Cardiomyopathy; ECG; Heart failure; Peripartum; Pregnancy.

Detecting cardiomyopathies in pregnant and postpartum women. AUC, area under the receiver operating characteristic curve; ECG, electrocardiogram.

Figure 1

Study population flow diagram. AV, atrioventricular; ECG, electrocardiogram; EF, ejection fraction.

Figure 2

(A–F) Receiver operating characteristic curves and confusion matrices for identification of cardiomyopathy among pregnant and postpartum women at pre-specified ejection fraction values. (A and B) Ejection fraction ≤35%; (C and D) ejection fraction <45%; and (E and F) ejection fraction <50%. AUC, area under the receiver operating characteristic curve.

Figure 3

Forest plots showing deep learning model performance for identification of left ventricular systolic dysfunction stratified by subgroups. (A) Ejection fraction ≤35%; (B) ejection fraction <45%; and (C) ejection fraction <50%. AUC, area under the receiver operating characteristic curve.

Figure 4

Receiver operating characteristic curves for identification of left ventricular ejection fraction <35% among pregnant and postpartum women who had NT-ProBNP Measured. AI, artificial intelligence; NT-ProBNP, N-terminal pro B-type natriuretic peptide.