Diller GP, Orwat S, Vahle J, et al. - Among 372 patients with tetralogy of Fallot (ToF) who had undergone cardiac magnetic resonance (CMR) imaging, researchers performed this inquiry to determine whether machine learning algorithms have utility for automatically estimating prognosis in these patients. They derived measures of cardiac dimensions and function by retrieving cine loops and subjecting these to automatic deep learning (DL)-based image analysis, trained on independent, local CMR data. In univariate Cox analysis, the outcome was shown to be significantly predicted by various DL parameters, including right atrial median area and right ventricular long-axis strain. Experts identified a link of DL parameters with adverse result independently of left and right ventricular ejection fraction and peak oxygen uptake. Through this work, data are afforded on the utility of machine learning algorithms trained on external imaging datasets to automatically estimate prognosis in patients with ToF. Because of the automated analysis process, these two-dimensional-based algorithms may afford substitutes for labour-intensive manually achieved imaging parameters among patients with ToF.