Multi-Label Cardiac Abnormality Classification from Electrocardiogram Using Deep Convolutional Neural Networks

Abstract

This paper proposes a deep neural network architecture to perform multi-label classification of 26 cardiac abnormalities from 12-lead and reduced lead ECG data. The model was created by team “NIMA” for the PhysioNet/Computing in Cardiology Challenge 2021. ECG signals of at most 20 seconds in length were used for training. The data are preprocessed by normalizing, resampling, and zero-padding to get a constant-sized array. The preprocessed ECG signals and Fast Fourier Transforms obtained from the preprocessed signals are each fed into two separate deep Convolutional Neural Networks. Spatial dropouts and average pooling are used between each convolutional layer to reduce overfitting and to reduce model complexity. Following the convolutional layers, the time and frequency domain network outputs are concatenated and passed through two dense layers that output an array of size 26. A threshold of 0.13 is used on the output array to determine the class while addressing data imbalance. The method achieved a score of 0.55, 0.51, 0.56, 0.55, and 0.56 ranking 2nd, 5th, 3rd, 3rd and 3rd out of 39 officially ranked teams on 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead hidden test datasets, respectively, according to the challenge evaluation metric. Our model performs comparable to the 12 Lead ECG using smaller subsets of leads.