Bouzid Z, Faramand Z, Gregg RE, et al. - In view of a limited sensitivity of classical ST‐T waveform changes on standard 12‐lead ECG in detecting acute coronary syndrome (ACS) in the emergency department and the uncertainty regarding the clinical utility of the various novel ECG features proposed to augment clinicians' decision during patient evaluation, researchers conducted this observational study of consecutive patients examined for suspected ACS (Cohort 1 n = 745, age 59±17, 42% female, 15% ACS; Cohort 2 n = 499, age 59±16, 49% female, 18% ACS). Out of 554 temporal‐spatial ECG waveform features, a subset of 65 physiology‐driven features that are mechanistically correlated with myocardial ischemia was selected using domain knowledge and their performance was compared to a subset of 229 data‐driven features selected by multiple machine learning algorithms. A final subset of 73 most important ECG features that had both data‐ and physiology‐driven basis to ACS prediction was then selected using random forest and their performance was compared with clinical experts. The subset of novel ECG features was identified as predictive of ACS with a fully interpretable model highly adaptable to clinical decision support applications. The improved reclassification performance was significantly attributable to metrics of nondipolar electrical dispersion (ie, circumferential ischemia), ventricular activation time (ie, transmural conduction delays), QRS and T axes and angles (ie, global remodeling), and principal component analysis ratio of ECG waveforms (ie, regional heterogeneity).