Advanced Techniques for The Elimination of EOG Artifacts in EEG Signals Employing CSSA Alongside Dynamic Wavelet Approaches

As for the analysis and interpretation of the electric discharges coming from the brain, EEG contamination is a rather big problem, especially in neurosciences, cognitive sciences, and diagnostics. The following presents a new technique for the elimination of artifacts from the EEG signal employed in this study. CSSA and DWT are employed for the purpose in the given approach. In SS EA method, the entire EEG signal is divided into segments and by using CSSA, the actual cerebral activity signals and the particular form of noise are differentiated. Thanks to the division of the signal into sub-bands with many resolutions and through the decision of the features of DWT it can reject the unwanted components which in any way has no relation to the brain structure. Using CSSA’s global decomposition and DWT’s localized time-frequency analysis, the hybrid technique guarantees that all of the EOG artifacts are deconvoluted while preserving all of the inherent neuronal information. The outcomes of the conducted experiments show that proposed method performs artifact removal and signal quality better than ICA methods. The improved data quality is characterised using real EEG data, which shows that the performance can be utilised in all fields where the potential of EEG is widely utilised.