Two Might Do: A Beat-by-Beat Classification of Cardiac Abnormalities using Deep Learning with Domain-Specific Features

Berken Utku Demirel1, Adnan Harun Dogan2, Mohammad Abdullah Al Faruque3
1University of California Irvine, 2Middle East Technical University, 3University of California, Irvine


This paper proposes an efficient convolutional neural network to detect 26 different classes of cardiac activities from different numbers of leads in the Physionet/Computing data in the Cardiology Challenge 2021. The proposed CNN architecture is designed to utilize heart rate variation features from ECG recordings and waveform morphologies of heartbeats simultaneously. Also, the designed architecture is flexible for the implementation of a different number of leads with a varied length of ECG recordings. The proposed algorithm achieved a score of 0.57 using only 2 channels of ECG on all recordings for the hidden validation set of the challenge, placing us 81, 64, 72, 71, 63rd (Team name: METU_19) out of 252 entries for 12, 6, 4, 3, and 2-leads respectively. These results show the potential of an efficient, flexible novel neural network for beat-by-beat classification of raw ECG recordings to a complex multi-class multi-label classification problem.