Robust arrhythmia classifier using wavelet transform and support vector machine classification

Abstract

The Electrocardiogram (ECG) is the most widely used signal in clinical practice for the assessment of cardiac condition. This paper presents a robust arrhythmia classifier based on the combination of wavelet transform and timing features, as well as support vector machine classification technique. The proposed technique is able to detect a total of 11 different types of arrhythmia. Results show that the average classification accuracy is up to 87.93% using the 46 MIT-BIH offline ECG database as the testing dataset. A user-friendly Graphical User Interface (GUI) is developed to ease the layman users. This proposed tool aims to reduce the workload of cardiac vascular technologist, medical staff and physicians as assisting cardiac monitoring equipment.