Abstract:

BACKGROUND: Fetal electrocardiogram is an important parameter in perinatal period for fetus growing. How to extract the fetal electrocardiogram from maternal electrocardiogram quickly and effectively has become an important research topic.
OBJECTIVE: To use the block-extend Infomax algorithm for the separation of the fetal electrocardiogram quickly and effectively.
METHODS: Block-extened Infomax and statistics into four ranks were combined to study the relate rules which make the non-cross element of weight covariance matrix minimum, and improve the astringency of arithmetic.
RESULTS AND CONCLUSION: Extended infomax algorithm could achieve the separation of super-Gaussian and sub-Gaussian signal. The convergence rate of the algorithm presented in the paper was faster than that of the extended infomax algorithm. Extended Infomax algorithm could effectively extract the fetal electrocardiogram. The amount of the data treated by the block-extend Infomax algorithm was less than that treated by the extended infomax algorithm, and the iteration number of the block-extend Infomax algorithm was smaller than that of the extended infomax algorithm, so the fetal electrocardiogram could be separated quickly and effectively.