A Novel Continuous Left Ventricular Diastolic Function Score Using Machine Learning

Abstract

Unlike left ventricular (LV) ejection fraction, which provides a precise, reliable, and prognostically valuable measure of systolic function, there is no single analogous measure of LV diastolic function. We aimed to develop a continuous score to grade LV diastolic function using machine learning modeling of echocardiographic data. Consecutive echo studies performed at a tertiary-care center between February 1, 2010, and March 31, 2016, were assessed, excluding studies containing features that would interfere with diastolic function assessment as well as studies in which 1 or more parameters within the contemporary diastolic function assessment algorithm were not reported. Diastolic function was graded based on 2016 American Society of Echocardiography (ASE)/European Association of Cardiovascular Imaging (EACVI) guidelines, excluding indeterminate studies. Machine learning models were trained (support vector machine [SVM], decision tree [DT], XGBoost [XGB], and dense neural network [DNN]) to classify studies within the training set by diastolic dysfunction severity, blinded to the ASE/EACVI classification. The DNN model was retrained to generate a regression model (R-DNN) to predict a continuous LV diastolic function score. A total of 28,986 studies were included; 23,188 studies were used to train the models, and 5,798 studies were used for validation. The models were able to reclassify studies with high agreement to the ASE/EACVI algorithm (SVM, 83%; DT, 100%; XGB, 100%; DNN, 98%). The continuous diastolic function score corresponded well with ASE/EACVI guidelines, with scores of 1.00±0.01 for studies with normal function and 0.74±0.05, 0.51±0.06, and 0.27±0.11 for mild, moderate, and severe diastolic dysfunction, respectively (mean±1 SD). A score of <0.91 predicted abnormal diastolic function (area under the receiver operator curve=0.99), while a score of <0.65 predicted elevated filling pressure (area under the receiver operator curve=0.99). Machine learning can assimilate echocardiographic data and generate an automated continuous diastolic function score that corresponds well with current diastolic function grading recommendations.