The Method of Independent Component Analysis in Removing Blink Artifacts from EEG
Xin Zhu · Journal of Tianjin University Science and Technology · 2003
The comparison between the Extended Infomax ICA and SVD methods in efficiency in extracting blink artifacts in the electroencephalograph(EEG) signals indicates that ICA algorithms are more insensitive to disturbance even in the conditions of low signal-noise-ratio than the commonly used SVD method. By applying Extended Infomax ICA, the blink artifacts contained in the 16-channel EEG signals of one Alzheimer-disease patient were removed successfully. ICA shows a promising prospect and important application value in the Biomedical Engineering, especially in the Clinic Medicine Engineering.