应用独立分量分析的胎儿心电信号提取

针对独立分量分析(independent component analysis,ICA)的基本生成模型,在采用负熵的快速定点ICA算法的基础上,应用渐近正交化的FastlCA算法,对3导从孕妇腹部测得的含有孕妇及其胎儿心电分量的观测信号进行了有效的分离,提取出胎儿心电分量;通过调用Matlab小波消噪程序,对分离所得的胎儿心电分量用db2小波对其做8层分解,获取默认软闽值,消噪处理.结果表明,渐近正交化的FastICA算法收敛速度快,只经过7、3、2次的迭代,便将3个源分量分离出来;结合小波阈值消噪,将分离后胎儿心电中的干扰进一步去除.

Extracting a fetal electrocardiogram based on independent component analysis

Based on the basic generation model for the independent component analysis(ICA) and the fixed-point FastICA algorithm using negentropy,three observed signals containing maternal electrocardiogram(MECG) and fetal electrocardiogram(FECG) components obtained from abdomen of a pregnant woman are successfully separated by using the FastICA algorithm with progressive orthogonalization,and the FECG is extracted.By invoking the wavelet de-noising program in Matlab,the extracted FECG is decomposed to 8 levels with d...