Bae Y, et al. - Given how challenging detection of vulnerable plaques by intravascular ultrasound (IVUS) is, researchers intended to create machine learning (ML) models that could predict optical coherence tomography-derived thin-cap fibroatheromas (OCT-TCFAs) using training vs test sets comprising 414 and 103 coronary lesions, respectively, in 517 patients with angina. They used 17 of 1,449 computed IVUS features, based on two-dimensional geometry and texture, in supervised ML with artificial neural network (ANN), support vector machine and naïve Bayes. The overall accuracies were demonstrated by ANN and naïve Bayes in the test set. Overall, the prediction of the presence of OCT-TCFA was enabled by supervised ML algorithms with computed IVUS features.