Attia ZI, Noseworthy PA, Lopez-Jimenez F, et al. - Researchers sought to identify the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm using standard 10-second, 12-lead ECGs, in order to create an artificial intelligence (AI)-enabled ECG using a convolutional neural network. They included 180,922 patients (aged 18 years or older) with 649,931 normal sinus rhythm ECGs in this work; eligible patients were those with at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the Mayo Clinic ECG laboratory. Patients with at least one ECG with a rhythm of atrial fibrillation or atrial flutter were classified as positive for atrial fibrillation. From 126,526 patients in the training dataset, 454,789 ECGs were recorded; from 18,116 patients in the internal validation dataset, 64,340 ECGs were recorded; and from 36,280 patients in the testing dataset, 130,802 ECGs were recorded. They identified verified atrial fibrillation before the normal sinus rhythm ECG tested by the model in 3,051 (8·4%) patients in the testing dataset. Findings support the utility of the AI-enabled ECG acquired during normal sinus rhythm in identification at the point of care of individuals with atrial fibrillation.