Analysis and Interpretation of ECG Time Series Through Convolutional Neural Networks in Brugada Syndrome Diagnosis

In this research, we present a novel approach to evaluate and interpret Convolutional Neural Networks (CNNs) for the diagnosis of Brugada Syndrome (BrS), a rare heart rhythm disease, from the electrocardiogram (ECG) time series. First, the model is assessed on the ECG classification of type-1 BrS. Then, we define a method to interpret the BrS prediction through Gradient-weighted Class Activation Mapping (Grad-CAM) applied to continuous time series. Finally, the proposed approach provides a tool to analyze the main areas of the ECG time series responsible for the BrS diagnosis through CNNs. In experimental assessments we use an original dataset of 306 ECGs collected from several clinical centers within the BrAID (Brugada syndrome and Artificial Intelligence applications to Diagnosis) project.