Baseline Wander Estimation for ECG Characterization

Abstract

Baseline wander makes ECG characterization difficult since it makes manual and automatic analysis of ECG difficult, especially the measuring of ST-segment deviation. The removing of baseline wander from ECG can cause distortion of important clinical information since the spectrum of baseline wander and low frequency component of ECG signal usually overlaps. The aim of this study is to use wavelet packet analysis for estimation of ECG baseline wander. The main reasons for using wavelet transform are the properties of good representation non-stationary signal such as ECG signal and the possibility of dividing the signal into different bands of frequency. To test the propose method, ECG signals taken from MIT-BIH arrhythmia database are used. The method had been compared with traditional methods such as moving average, median etc. on the artificially constructed ECG with baseline wander. For the performance analysis we have use Percent Root Mean Square Difference (PRD). Obtained results show that the wavelet transform approach performs better than traditional methods.