Lee JG, Ko J, Hae H, et al. - Given that the functional significance of intermediate coronary stenosis is poorly predicted by intravascular ultrasound (IVUS)-derived morphological criteria, researchers focused on the diagnostic performance of IVUS-based supervised machine learning (ML) algorithms in detecting lesions with a fractional flow reserve ≤ 0.80 (vs > 0.80). This study involved 1,328 patients with 1,328 non-left main coronary lesions. The participants were randomized into training and test sets in a 4:1 ratio. Using an automatic segmentation model, masked IVUS images were produced. For ML training with 5-fold cross-validation, they used 99 computed IVUS features and six clinical variables (age, gender, body surface area, vessel type, involved segment, and involvement of the proximal left anterior descending artery). They used the non-overlapping test samples to assess binary classifiers with respect to their diagnostic performances for identifying ischemia-producing lesions. Findings revealed the good diagnostic performance of the IVUS-based ML algorithms for detecting ischemia-producing lesions and also their possible capability to decrease the requirement for pressure wires.