Classification of Arrhythmia by Using Deep Learning with 2-D ECG Spectral Image Representation

Amin Ullah,Raja Majid Mehmood,Syed Muhammad Anwar,Muhammad Bilal

doi:10.3390/rs12101685

Abstract

The electrocardiogram (ECG) is one of the most extensively employed signals used in the diagnosis and prediction of cardiovascular diseases (CVDs). The ECG signals can capture the heart’s rhythmic irregularities, commonly known as arrhythmias. A careful study of ECG signals is crucial for precise diagnoses of patients’ acute and chronic heart conditions. In this study, we propose a two-dimensional (2-D) convolutional neural network (CNN) model for the classification of ECG signals into eight classes; namely, normal beat, premature ventricular contraction beat, paced beat, right bundle branch block beat, left bundle branch block beat, atrial premature contraction beat, ventricular flutter wave beat, and ventricular escape beat. The one-dimensional ECG time series signals are transformed into 2-D spectrograms through short-time Fourier transform. The 2-D CNN model consisting of four convolutional layers and four pooling layers is designed for extracting robust features from the input spectrograms. Our proposed methodology is evaluated on a publicly available MIT-BIH arrhythmia dataset. We achieved a state-of-the-art average classification accuracy of 99.11%, which is better than those of recently reported results in classifying similar types of arrhythmias. The performance is significant in other indices as well, including sensitivity and specificity, which indicates the success of the proposed method.

Highlights

Cardiovascular diseases (CVDs) are the leading cause of human death, with over 17 million people known to lose their lives annually due to CVDs [1]
We used 1-D ECG signals as input to the convolutional neural network (CNN) model used in experiments and achieved a classification accuracy of 97.80%
One of the reasons for these observations could be the deeper architecture of VGGNet and limited training data. These results prove that the proposed CNN model has the state-of-the-art accuracy for the automatic classification of arrhythmia based on the comparison with different CNN based algorithms

Summary

Introduction

Cardiovascular diseases (CVDs) are the leading cause of human death, with over 17 million people known to lose their lives annually due to CVDs [1]. One of the common sources of CVDs is cardiac arrhythmia, where heartbeats are known to deviate from their regular beating pattern. In cases where the heartbeat feels too fast or slow, the condition is known as palpitations. An arrhythmia does not necessarily mean that the heart is beating too fast or slow, it indicates that the heart is following an irregular beating pattern. Some other common types of abnormal heart rhythms include atrial fibrillation, atrial flutter, and ventricular fibrillation. These deviations could be classified into various subclasses and represent different types of cardiac arrhythmia. Arrhythmia could either mean a slow or fast beating of heart, or patterns that are not attributed to a normal heartbeat. There are certain known characteristics of cardiac arrhythmia, where the detection requires expert clinical knowledge

Methods

Results

Discussion

Conclusion