CLASSIFICATION OF CARDIAC ARRHYTHMIAS WITH ARTIFICIAL NEURAL NETWORKS ACCORDING TO GENDER DIFFERENCES

Abstract

Cardiac arrhythmias are common heart diseases. Electrocardiography (ECG) is an important measure for diagnosing arrhythmias. Researchers use the ECG signals in order to train artificial neural networks (ANN). In previous studies the ECG signals of males and females were analysed together. We know that there are some differences between male and female ECG signals. This paper suggests that classifying the arrhythmias according to gender differences gives more accurate results. In this study we classify the subjects as normal and right bundle branch block (RBBB) using cascade forward back algorithm in MATLAB. The accuracy of network simulations are as follow: 81.25% only male, 80% only female, 40% male and female together.

Keywords

• Arrhythmia
• Cascade forward back propagation
• Gender differences
• MATLAB