ECG baseline wander correction by mean-median filter and discrete wavelet transform

Abstract

Electrocardiographic (ECG) analysis plays an important role in diagnosis of heart diseases. High quality ECG pushes forward new drug development and improves clinical diagnosis. This paper introduces a novel method to correct baseline wander (BW) components of ECG signals based on Mean-Median (MEM) filter and discrete wavelet transform (DWT). We obtain the BW estimation via MEM, and decompose the estimation into different scales by DWT. Then, an iterative sifting process based on t-test is adopted to select the scales to reconstruct the refined BW components. The proposed method is applied to MIT-BIH Arrhythmia Database. The experimental results verify that the proposed method can effectively remove BW components and preserve useful waveform information.