TABLE 3.

Performance of machine learning data-driven approaches, with methods (a) support vector machines (SVM) using radial basis function (RBF) kernel and (b) decision tree, for different combinations of the optical flow, rate of change of myocardial area and radiomics data.The SVM model had the best performance in terms of accuracy and F1 scores when the input consists of the rate of change of myocardial area and the radiomics data. The DT model had the best in terms of the precision score when the input consists of the optical flow, the rate of change of myocardial area and the radiomics data.