12-lead ECG Arrythmia Classification Using Convolutional Neural Network for Mutually Non-Exclusive Classes

Abstract

The growing demand for diagnosing of cardiovascular diseases leads to the development of new solutions for automatic classification of recorded ECG signals. Creating a robust and fast algorithm for automatic classification of ECG signal is crucial to improve the quality of healthcare, especially in countries where a lack of experienced specialists is an issue or the healthcare system is overloaded. The aim of the PhysioNet/Computing in Cardiology Challenge 2020 is to create an algorithm for classification of 12-lead ECGs based on ECG signals from multiple databases across the world. The shared training set consisted of 43,101 ECG recordings lasting from 5 to 1800 seconds. We (BioS Team) proposed the machine learning algorithm based on convolutional neural networks. The ECG signals were pre-processed using moving median filters to remove high-frequency noise and baseline wandering. We developed simply convolutional neural network consisting of four main convolutional blocks and one fully connected layer. We achieved a challenge validation score of 0.349, and full test score of 0.279, placing us 14 out of 41 in the official ranking.