Data Science Modeling for EEG Signal Filtering Using Wavelet Transforms
Ivan G. Garvanov, Vladimir Jotsov, Magdalena Z. Garvanova · 2020
Modeling and constraint satisfaction applications had been considered aiming at diminishing the manual labor during the wavelet signal filtering. Electroencephalogram (EEG) signals are easily affected by various noise sources. The noise can be electrode noise or can be generated from the body itself. The noises in the EEG signals are called artifacts and these artifacts are needed to be removed from the original EEG signals for the proper analysis of the signals. This work presents denoising algorithm based on the combination of wavelet transform (WT), threshold processing and inverse wavelet transform. The proposed algorithm is tested using real EEG signals. To improve its efficiency, different modeling and data preprocessing methods had been applied.