Ahnaf Shahrear Khan & Mamun Miah - Computer Science & Engineering, University of Rajshahi
EEG recordings are usually affected by various types of artifacts that come from non-neural sources and make it harder for accurate signal classification in later stages. So, reliably detecting and removing artifacts from EEG signals by an automated signal processing algorithm is an active research area. In this paper, we have developed a wavelet-based artifacts suppression algorithm for EEG signals that selects the best optimal threshold parameters, and hence consequently provides the best performance of artifact removal. In the proposed algorithms, we choose to sweep the threshold parameters until the best accuracy or least distortion is achieved by making a decision based on a reference dataset. The criteria for optimized selection are based on matrices that quantify the ratio between the signal and the noise, mean square error, etc. The algorithm is tested on a real dataset of EEG signals that have ocular artifacts. The achieved results prove that selecting the optimum threshold parameter values adaptively would give the best performance compared with selecting any predefined threshold parameters. This research would help the EEG signal analysis community with a platform to work further in the future on such problem to be able to properly select the wavelet parameters.
