A review of electrocardiogram filtering

Abstract

Analog filtering and digital signal processing algorithms in the preprocessing modules of an electrocardiographic device play a pivotal role in providing high-quality electrocardiogram (ECG) signals for analysis, interpretation, and presentation (display, printout, and storage). In this article, issues relating to inaccuracy of ECG preprocessing filters are investigated in the context of facilitating efficient ECG interpretation and diagnosis. The discussion covers 4 specific ECG preprocessing applications: anti-aliasing and upper-frequency cutoff, baseline wander suppression and lower-frequency cutoff, line frequency rejection, and muscle artifact reduction. Issues discussed include linear phase, aliasing, distortion, ringing, and attenuation of desired ECG signals. Due to the overlapping power spectrum of signal and noise in acquired ECG data, frequency selective filters must seek a delicate balance between noise removal and deformation of the desired signal. Most importantly, the filtering output should not adversely impact subsequent diagnosis and interpretation. Based on these discussions, several suggestions are made to improve and update existing ECG data preprocessing standards and guidelines.