A Deep Learning Approach for Detecting Atrial Fibrillation using RR Intervals of ECG

Abstract

Atrial Fibrillation (AF) is one of the most common type of heart arrhythmias observed in the clinical practice. AF can be detected using Electrocardiogram (ECG). ECG signal are time varying and nonlinear in nature. Hence, it is very difficult for a physician to perform accurate and rapid classification of different heart rhythms, manually. In this paper we propose a method using Discrete Wavelet Transform (DWT) with db6 as basis function for denoising ECG signal along with Savitzky- Golay filter to smoothen the signal. Deep learning methods, such as combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) (CNN-LSTM) and ResNet18 are used for accurate classification of ECG signal using Physionet Challenge 2017 database. With 10-fold cross validation method the model provided overall accuracy of 98.25% with CNN-LSTM classifier.

Keywords

• Atrial Fibrillation
• Electrocardiogram
• Discrete Wavelet Transform
• Savitzky-Golay Filter
• Convolutional Neural Network
• Long Short Term Memory