Noninvasive fetal electrocardiogram extraction: Blind separation versus adaptive noise cancellation

被引：207

作者：

Zarzoso, V ^{[1
]}

Nandi, AK ^{[1
]}

机构：

[1] Univ Liverpool, Signal Proc & Commun Grp, Dept Elect Engn & Elect, Liverpool L69 3GJ, Merseyside, England

来源：

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING | 2001年 / 48卷 / 01期

关键词：

adaptive noise cancellation; blind source separation; fetal electrocardiogram extraction; higher-order statistics; independent component analysis; optimal Wiener-Hopf filtering;

D O I：

10.1109/10.900244

中图分类号：

R318 [生物医学工程];

学科分类号：

0831 ;

摘要：

The problem of the fetal electrocardiogram (FECG) extraction from maternal skin electrode measurements can be modeled from the perspective of blind source separation (BSS), Since no comparison between BSS techniques and other signal processing methods has been made, we compare a BSS procedure based on higher-order statistics and Widrow's multireference adaptive noise cancelling approach. As a best-case scenario for this latter method, optimal Wiener-Hopf solutions are considered. Both procedures are applied to real multichannel ECG recordings obtained from a pregnant woman. The experimental outcomes demonstrate the more robust performance of the blind technique and, in turn, verify the validity of the BSS model in this important biomedical application.

引用

页码：12 / 18

页数：7