Machine Learning Assessment of Left Ventricular Diastolic Function Based on Electrocardiographic Features

• Published in: Journal of the American College of Cardiology Vol. 76; no. 8; pp. 930 - 941
• Main Authors: Kagiyama, Nobuyuki, Piccirilli, Marco, Yanamala, Naveena, Shrestha, Sirish, Farjo, Peter D., Casaclang-Verzosa, Grace, Tarhuni, Wadea M., Nezarat, Negin, Budoff, Matthew J., Narula, Jagat, Sengupta, Partho P.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 25.08.2020
• Subjects: Cardiovascular; Early Diagnosis; echocardiography; Echocardiography - methods; electrocardiogram; Female; Heart Failure, Diastolic - diagnosis; Heart Failure, Diastolic - physiopathology; Humans; left ventricular diastolic dysfunction; Machine Learning; Male; Middle Aged; Myocardial Contraction - physiology; myocardial relaxation; Predictive Value of Tests; ROC Curve; Signal Processing, Computer-Assisted; Stroke Volume; Ventricular Dysfunction, Left - diagnosis; Ventricular Dysfunction, Left - physiopathology; spECG; echocardiography; ECG; LVEF; LVDD; LV; AUC; machine-learning; electrocardiogram; left ventricular diastolic dysfunction; NPV; PPV; myocardial relaxation; TR; positive predictive value; signal-processed electrocardiogram; area under the curve; left ventricular ejection fraction; tricuspid regurgitation; left ventricle/ventricular; negative predictive value
• ISSN: 0735-1097; 1558-3597; 1558-3597
• DOI: 10.1016/j.jacc.2020.06.061

Left ventricular (LV) diastolic dysfunction is recognized as playing a major role in the pathophysiology of heart failure; however, clinical tools for identifying diastolic dysfunction before echocardiography remain imprecise. This study sought to develop machine-learning models that quantitatively estimate myocardial relaxation using clinical and electrocardiography (ECG) variables as a first step in the detection of LV diastolic dysfunction. A multicenter prospective study was conducted at 4 institutions in North America enrolling a total of 1,202 subjects. Patients from 3 institutions (n = 814) formed an internal cohort and were randomly divided into training and internal test sets (80:20). Machine-learning models were developed using signal-processed ECG, traditional ECG, and clinical features and were tested using the test set. Data from the fourth institution was reserved as an external test set (n = 388) to evaluate the model generalizability. Despite diversity in subjects, the machine-learning model predicted the quantitative values of the LV relaxation velocities (e’) measured by echocardiography in both internal and external test sets (mean absolute error: 1.46 and 1.93 cm/s; adjusted R2 = 0.57 and 0.46, respectively). Analysis of the area under the receiver operating characteristic curve (AUC) revealed that the estimated eʹ discriminated the guideline-recommended thresholds for abnormal myocardial relaxation and diastolic and systolic dysfunction (LV ejection fraction) the internal (area under the curve [AUC]: 0.83, 0.76, and 0.75) and external test sets (0.84, 0.80, and 0.81), respectively. Moreover, the estimated eʹ allowed prediction of LV diastolic dysfunction based on multiple age- and sex-adjusted reference limits (AUC: 0.88 and 0.94 in the internal and external sets, respectively). A quantitative prediction of myocardial relaxation can be performed using easily obtained clinical and ECG features. This cost-effective strategy may be a valuable first clinical step for assessing the presence of LV dysfunction and may potentially aid in the early diagnosis and management of heart failure patients. [Display omitted]