Figure 3.

Overview of the deep learning architecture. (A) For the primary output, 206 ECGs were loaded into the convolutional neural network model and classified into four classifications (Group A, Group B, Group C, and Group N [normal class]). (B) For the secondary output, chest X-ray images from 1519 patients were used for pretraining, and the original chest X-ray images were compressed into a one-dimensional vector using the pretrained midlayer weights. The combined ECG and chest X-ray data were finally trained to classify the outputs. 2D-CNN indicates a two-dimensional convolutional neural network, and 1D-CNN indicates a one-dimensional convolutional neural network.