Development of arrhythmia classification system for personal cardiac monitor in thailand

Abstract

This research has developed an arrhythmia detection system for the person who has a cardiovascular disease based electrocardiography signals. The heart disease is the leading cause of death in Thailand such as a heart failure and stroke. The purpose of an arrhythmia detection system is to classify a normal sinus rhythm and three types of arrhythmia rhythm (Atrial fibrillation, Bradycardia and Complete heart block) with a high accuracy rate. The prototype of our system, we used ECG signal from the standard vital sign simulator. The well-known algorithm, namely Pan-Tomkins Algorithm used to detect QRS complex and calculate EMG features are the mean and standard deviation of the RR interval. The output is classified by using Support Vector Machine. The accuracy of our system can discriminate between non-artifact of 95.83% and artifact with 25% of 96.43%. The result is very promising. © 2019 IEEE.