Power line and ocular artifact denoising from EEG using notch filter and wavelet transform

Electroencephalogram (EEG) records deliver the information about anomaly and/or results to some impetus in the human brain. However, these signals get normally mixed with other biomedical signals. For example, theta and alpha brain rhythm is often mixed with the EOG. The eye blinks and vertical/horizontal/ round movements of the eyes develop artifacts in the EEG. Other artifact sources are the ECG, EMG, and the power line artifact (50 Hz). The presence of artifacts makes the EEG analysis difficult and noises introduce spikes/transients etc. with high amplitude that may mislead with neurological rhythms. Thus, the undesirable signals should be phase out from the EEG to assure an appropriate analysis and diagnosis. Conventional filtering cannot be employed to wipe out these types of OAs because EEG signal and artifacts have overlapping spectra. Adaptive filters have the efficiency of correcting their properties according to selected features of the signals being analyzed. In this study, a method, SURE adaptive thresholding of wavelet coefficients is used to remove frequent OA and IIR notch filter of 50 Hz to remove power line interference from EEG signals. The prime objective of this work intends to absolutely remove the power line artifacts and EOG from EEG by preserving the original cerebral signals.