Hybrid Arrhythmia Detection on Varying-Dimensional Electrocardiography: Combining Deep Neural Networks and Clinical Rules

Abstract

Aim: This study (from Revenger team) aims to develop effective approaches for the detection of cardiac arrhythmias from varying-dimensional electrocardiography (ECG) in the PhysioNet/Computing in Cardiology Challenge 2021, taking advantage of both deep neural networks (DNNs) and insights from clinical diagnostic criteria. Methods: 26 classes (equivalent classes are counted one) of ECGs are divided into two categories. Detectors are manually designed for classes in the category with clear clinical rules. The rest classes with subtle morphological and spectral characteristics are classified by DNNs. To make the networks capable of capturing features of different scopes, we use multi-branch convolutional neural networks (CNNs), each with different receptive fields via dilated convolutions. Considering ECGs' varying dimensionality, convolutions are grouped with group number equaling the number of leads. Outputs from DNNs and from manual detectors are merged to give final predictions. Results: Although we did not officially rank (the code failed to complete on the 12-lead test set), we received test scores of 0.33, 0.35, 0.33, 0.33, and 0.33 on the 2-lead, 3-lead, 4-lead and 6-lead test sets respectively. Conclusion: The proposed hybrid method is effective for establishing auxiliary diagnosis systems, and the reduced-lead ECGs are sufficient for such systems.