R-Peak Detection in Holter ECG Signals Using Non-Negative Matrix Factorization

Abstract

Holter monitoring is mainly used for medical followup and diagnosis of patients with suspected cardiac arrhythmia such as heart rhythm irregularities that can be missed during classical electrocardiogram recording (ECG). However, these long-term continuous recordings represent a large amount of data that cannot be processed by hand. In this article, we present a new method based on Non-negative Matrix Factorization (NMF) to detect R-peaks in Holter signals. The approach consists in two stages: source separation based on the different time-frequency patterns of the QRS complexes and the other waves of the signal (P and T waves) and R-peak detection using Automatic Objective Thresholding (AOT). The proposed approach is validated on the MIT-BIH Arrhythmia database and achieves an average sensitivity of 99.59% and a precision of 99.69%. Using the MIT-BIH Noise Stress Test database, we also show the ability of our approach to discriminate R-peaks in signals contaminated with different noises.