Towards Generalization of Cardiac Abnormality Classification Using ECG Signal

Xiaoyu Li1, Chen Li1, Xian Xu2, Yuhua Wei1, Jishang Wei3, Yuyao Sun2, Buyue Qian4, Xiao Xu2
1Xi'an Jiaotong University, 2Ping An Health Technology, 3HP Labs, 4The First Affiliated Hospital of Xi'an Jiaotong University


In the PhysioNet/Computing in Cardiology Challenge 2021, our team, DrCubic, develops a novel approach to classify cardiac abnormalities using reduced-lead ECG recordings. In our approach, we incorporate peak detection as a self-supervised auxiliary task. We build the model based on SE-ResNet, and ensemble models of different input lengths and sampling rates. Inspired by last year's challenge results, we investigate various settings and techniques, and select the best ones, considering the intra-source performance and inter-source generalization simultaneously. Our classifiers receive scores of 0.666, 0.643, 0.642, 0.651, and 0.639 (ranked 3rd, 3rd, 4th, 4th, and 3th out of the teams) for the 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead versions of the hidden validation set with the Challenge evaluation metric.