Abstracto

Fetal electrocardiogram extraction combining with wavelet de-noising and blind source separation algorithm

Chen Yu, Wen Xinling, Liu Zhaoyu

In practical application, the weak signal of fetal electrocardiogram (FECG) is obtained through the surface electrodes placed upon the maternal abdominal with all kinds of other signals, such as the maternal electrocardiogram (MECG), work frequency jamming, and noise signal, etc, therefore, the application of FECG extraction algorithm in noise mixture circumstance has the practical significance and important clinical value. Independent Component Analysis (ICA) is a new blind source separation technology developed in recent years. And blind source separation (BSS) algorithm based on ICA technology has been widely applied to the medical field. This paper adopted a kind of popular BSS algorithm of Fast ICA combining with the wavelet de-noising algorithm to separate the FECG signal. Through the experiment, the MECG, FECG and noise can be separation from the mixture signal, and after the de-noising to the FECG signal, we can extract more clear signal effectively