Automatic removal of ocular and cardiac artifacts and retention of embedded EEG information via adaptive noise cancellation

Abstract

Electroencephalography (EEG) reflects the spontaneous electrophysiological activity in the cerebral cortex of the brain. In practice, these recordings are often contaminated by the artifacts. Removal of these artifacts is necessary before correct processing of the EEG signal. In this study, automatic physiological artifacts removal from the EEG signal is attained by commonly used method, Independent Component Analysis (ICA). The inability of ICA to keep intact the original EEG information after the removal of physiological artifacts is highlighted using two parameters: Correlation coefficient and Signal to Noise Ratio (SNR). In addition, an alternative method based on Adaptive Noise Cancellation (ANC) for EEG signal retention after artifact removal is proposed. The experimental results suggest that ANC is more suited for artifact removal as compared to ICA. It is inferred that ANC is a better method than ICA in retaining the EEG signal in the non-artifactual region while automatically suppressing the artifactual activity.