A Nonlinear Digital Filter For Cardiac QRS Complex Detection

Publisher Website
Google Scholar
Abstract
Many of the usual methods for detection of QRS complexes in ECG analysis or monitoring systems have been based on linear filters. These filters may miss ectopic beats of small amplitudes, mistake tall T-waves as QRS complexes, and even smooth signal edges. In this report, an algorithm based on a nonlinear digital filter, which is a running filter with hanning window smoothing, is fully described. The purpose of this algorithm is to recognize almost all bizarre shapes of QRS waveforms in a population of 2,200 patients. Compared to three cardiologists' independent visual readings as the accepted standard, this algorithm proved to be simple, fast, and easy to implement, with an overall accuracy of 99.6% in the tests performed.

This publication has 0 references indexed in Scilit: