Separation performance of ICA algorithms on FECG and MECG signals contaminated by noise

Abstract

This paper evaluates the performance of some major ICA algorithms like Bell and Sejnowski's infomax algorithm, Cardoso's Joint Approximate Diagonalization of Eigen matrices (JADE) and Comon's algorithm in a biomedical blind source separation problem. Independent signals representing Fetal ECG (FECG) and Maternal ECG (MECG) are generated and then mixed linearly in the presence of white or pink noise to simulate a recording of electrocardiogram. ICA has been used to extract FECG, but very less literature is available on the performance, i. e. , how does it behave in clinical environment. So there is a used to evaluate performance of these algorithms in Biomédical. To quantify the performance of ICA algorithms, two scenarios, i. e. , (a) different amplitude ratios of simulated maternal and fetal ECG, (b) different values of additive white gaussian noise or pink noise, were investigated. Higher order and Second order performances were measured by performance index and signal-to-error ratio respectively. The selected ICA algorithms separate the white and pink noises equally well. The performance of the Comon's algorithm is slightly less compared to the other two algorithms. © Springer-Verlag 2004.