Research Paper Volume 15, Issue 9 pp 3524—3537
The feasibility of early detecting coronary artery disease using deep learning-based algorithm based on electrocardiography
Figure 5. The architecture of the proposed CNN model. We used ResNet-50 uniting a mixed “attention” module (the Squeeze-and-Excitation module) as backbone architecture to extract useful features from the input ECG raw data. Abbreviations: ConV: Convolution; BN: Batch Normalization; Relu: Rectified Linear Units.