Title: Performance comparison of genetic algorithm and principal component analysis methods for ECG signal extraction

Authors: S. Balambigai, R. Asokan

Addresses: Department of ECE, Kongu Engineering College, Perundurai – 638052, Erode Dt., Tamil Nadu, India. ' Department of IT, Kongu Engineering College, Perundurai – 638052, Erode Dt., Tamil Nadu, India

Abstract: Electrocardiogram (ECG) signal analysis is a technique to diagnose the cardiac diseases. But, the desired electrocardiogram signals are often corrupted by baseline interference, power line interference and electromyogram. Here, a method is proposed to extract ECG from noisy signals based on Singular Value Decomposition (SVD) and Genetic Algorithm. The advantage of this method compared to conventional methods like adaptive filtering, neural networks is that it does not require any prior knowledge of the signals. It is found that the signal to noise ratio improvement is nearly double when compared to neural network methods.

Keywords: genetic algorithms; GAs; ECG signal extraction; electrocardiograms; SVD; singular value decomposition; eigenvalues; signal to noise ratio; PCA; principal component analysis; ICA; independent component analysis; cardiac disease; heart disease; neural networks.

DOI: 10.1504/IJHTM.2011.042369

International Journal of Healthcare Technology and Management, 2011 Vol.12 No.5/6, pp.379 - 389

Published online: 28 Mar 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article