A global approach for automatic artifact removal for standard EEG record

Area of Science:

• Neuroscience
• Signal Processing
• Biomedical Engineering

Background:

• Electroencephalography (EEG) records brain activity via scalp electrodes.
• EEG signals are frequently contaminated by artifacts from various sources like eye blinks, eye movements, and muscle activity.
• Effective artifact removal is crucial for accurate EEG data analysis.

Purpose of the Study:

• To present a global artifact removal method for EEG signals.
• To utilize independent component analysis (ICA) applied to frequency-band-filtered signals.
• To automate the identification and filtering of artifactual sources in EEG data.

Main Methods:

• The study employs independent component analysis (ICA) for artifact detection and removal.
• EEG signals are segmented into different frequency bands prior to analysis.
• A global filtering approach using constant bases is applied to the identified artifactual components.

Main Results:

• The proposed ICA-based method demonstrates effective identification of artifactual sources.
• Global filtering using constant bases successfully removes artifacts from EEG signals.
• The method offers an automated solution for EEG artifact management.

Conclusions:

• The developed method provides an automated and efficient approach to EEG artifact removal.
• This technique enhances the reliability and accuracy of EEG data for research and clinical applications.
• Further validation of the method's performance across diverse EEG datasets is recommended.