Electrooculographic and electromyographic artifacts removal from EEG

摘要

Completely automated methods for removing electroocular (EOG) and muscle (EMG) artifacts in the electroencephalogram (EEG) are presented in this paper which are implemented as an Automatic Artifact Removal (AAR) toolbox in EEGLAB. The method of removing EOG artifacts is based on second order blind identification (SOBI) as a blind source separation (BSS) technique and procedure of muscle artifact removal is based on canonical correlation analysis (CCA) as a blind source separation (BSS) technique. These techniques successfully deals with the removal of the most EOG and EMG artifacts in EEG recordings containing epilectic seizures without distorting the recorded ictal activity.