A New Wavelet-based Algorithm for R-peak Detection in ECG and it’s a Comparison with the Currently Existing Algorithms

Abstract

The purpose of this paper is to develop a new approach for Rpeak detection in ECG and compare it with the most effective and existing algorithms. The proposed approach is based on DWT and envelope for the first stage of preprocessing and decision or detecting it was achieved by adaptive thresholding. The proposed algorithm is compared with Pan & Tompkins, Savitzky-Golay smoothing filter, Hilbert and wavelet transforms as well as fast Fourier transform algorithm to investigate these techniques of R peak detection and evaluate the wavelet-based algorithm comparing with them. The algorithms are evaluated in the experimental section using ECG signals from the MIT-BIH database. The results of detection algorithms show that the proposed wavelet-based algorithm gets the highest sensitivity by 99.9% with higher reliability compared to other algorithms, also by analyzing the precision of them, it's come to light that FFT improved the highest precision with 99.7%.

Pre-processing of ECG signal based on multistage filters

“…The frequency of the baseline wander noise ranges from 0.1 to 0.5 Hz. However, this noise complicates peak detection because, due to wander, the T peak is higher than the R peak, resulting in peak detection errors [11]. Furthermore, the ECG signal is heavily influenced by EMG noise due to direct interference with the QRS complex [12].…”