NN-Based R-peak Detection in QRS Complex of ECG Signal

Abstract

-eural Network (NN) is designed to detect QRS complex from ECG signal. QRS complex detection is essential so that RR-interval can be measured for disease classification and can also be monitoring the heart rate. In this paper, a supervised Neural Network based algorithm has been used to detect R in QRS complex. It was tried to find out the R-peak in QRS complex with missing peak and false peak as well, so that the correct decision can be made by the physician and clinician. The accuracy of finding the R-peak by using the Neural Network was 99.09% averagely and the average percentage of missing and false peak was 00.09%. The technique appears to be exceedingly robust, correctly detects R-peaks even aberrant QRS complexes in noise-corrupted ECGs.

A Robust R Peak Detection Algorithm Using Wavelet Transform for Heart Rate Variability Studies

“…These works suffer mainly from two limitations: their sensitivity to noise and the choice of a threshold. Other works exploit nonlinear analysis methods, especially the neural networks [5,6] and non-stationary analysis tools such as wavelets that are the most used [7,8,9]. Particularly in [7], for the QRS complex detection by Discreet Wavelet Transform, The authors have used different mother wavelets: Cubic Spline, Haar and Daubechies4 (Db4), where the choice of decomposition scale is empirical .They have shown that the best results are given by the Cubic Spline and the Db4 wavelets.…”