被撤回的出版物: Artifacts and noise removal in electrocardiograms using independent component analysis (Retracted Article)

[摘要] Independent component analysis (ICA) is a novel technique capable of separating independent components from electrocardiogram ( ECG) complex signals. The purpose of this analysis is to evaluate the effectiveness of ICA in removing artifacts and noise from ECG recordings. ICA is applied to remove artifacts and noise in ECG segments of either an individual ECG CSE data base file or all files. The reconstructed ECGs are compared with the original ECG signal. For the four special cases discussed, the R-Peak magnitudes of the CSE data base ECG waveforms before and after applying ICA are also found. In the results, it is shown that in most of the cases, the percentage error in reconstruction is very small. The results show that there is a significant improvement in signal quality, i.e. SNR. All the ECG recording cases dealt showed an improved ECG appearance after the use of ICA. This establishes the efficacy of ICA in elimination of noise and artifacts in electrocardiograms. (C) 2007 Elsevier Ireland Ltd. All rights reserved.

[发布日期] 2008-09-26

[效力级别]  Retracted Publication

[关键词] electrocardiogram;feature extraction;reconstruction error;kurtosis;variance of variance;modeling;independent component analysis