ECG Classification using Deep Convolutional Neural Networks and Data Analysis

Abstract

ECG (Electrocardiogram) is a reliable and efficient test for monitoring the activities inside the cardiovascular system. The ECG reports are used to measure the electrical activities of the heartbeat which can be useful for many important conclusions regarding the heart diseases. In the recent past, there has been a lot of attention for the classification of these heartbeats using the ECG reports. In the last few years, Artificial Intelligence and Machine Learning are serving a lot in the area of automation in the medical and health-care domain. Deep Learning techniques are really growing day-by-day and Neural Networks are one of the major advancements in it. Neural Networks are really efficient in the problems like classification and segmentation. This study proposes a Deep CNN based model for classification of heartbeat using the ECG reports in five different classes which correspond to the different types of arrhythmia that are according to the standards of AAMI-EC57. We have evaluated our model on the Physionet's MIT-BIH and PTB Diagnostics datasets.