Computer-Aided Model for Abnormality Detection in Biomedical ECG Signals

Abstract

The paper introduces a new model that integrates the wavelet packet transform and ECG signal feature extraction for effect abnormality detection in ECG signals. It presents a brief description of ECG signal and its characteristics. At first, Wavelet decomposition is used for analyzing ECG signals, and extracting some features in order to increase the reliability of QRS detection. Then, major components of the ECG signal such as P wave, QRS complex and T wave have been detected to extract some features. Finally, the beats have been classified to detect the cardiac problems known as arrhythmia including tachycardia and bradycardia. Some recordings of the MIT-BIH Arrhythmia Database have been used.