Automatic Eyeblink Artifact Removal From EEG Signal Using Wavelet Transform With Heuristically Optimized Threshold.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Neuroscience

Background:

• Electroencephalogram (EEG) signals are highly susceptible to artifacts, particularly from eye blinks, which can obscure underlying neural activity.
• Accurate artifact removal is crucial for reliable EEG data analysis in clinical and research settings.

Purpose of the Study:

• To develop an automated and effective method for removing eyeblink artifacts from corrupted EEG signals.
• To introduce a novel thresholding strategy for approximation coefficients in Discrete Wavelet Transform (DWT) for artifact removal.

Main Methods:

• EEG signals were first classified for eyeblink corruption using Support Vector Machine (SVM).
• Corrupted EEG signals were decomposed using DWT up to the sixth level.
• A novel backward thresholding of approximation coefficients (ACs) was applied, with optimal thresholds determined by Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO).
• Inverse DWT (IDWT) was used to reconstruct artifact-free EEG signals.

Main Results:

• The proposed method successfully identified and removed eyeblink artifacts from EEG signals.
• Reconstructed EEG signals showed superior quality compared to existing methods, as indicated by a higher average correlation coefficient (CC).
• The meta-heuristic optimization algorithms (PSO and GWO) effectively determined optimal threshold values.

Conclusions:

• The developed automatic method provides a superior and fully automatic approach for eyeblink artifact removal from EEG signals.
• Thresholding approximation coefficients in a backward manner using optimized thresholds is an effective strategy for enhancing EEG signal reconstruction.
• The proposed technique offers significant improvements in EEG signal quality for subsequent analysis.