Skip to main content
. 2022 Nov 2;3(4):654–657. doi: 10.1093/ehjdh/ztac065

Graphical Abstract.

Graphical Abstract

Convolutional neural network models were trained using training data at each institution. Then the models were tested in the held-out test set from the institution and in three external test sets from the remaining three institutions, and their performances were compared. AUROC, area under the receiver operating curve; BWH, Brigham and Women’s Hospital; CNN, convolutional neural network; Keio, Keio University Hospital; MGH, Massachusetts General Hospital; UCSF, University of California San Francisco.