Independent component analysis and multiresolution asymmetry ratio for brain-computer interface

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Brain-computer interfaces (BCI) enable communication and control through brain signals.
• Electroencephalogram (EEG) data is crucial for BCI but susceptible to artifacts like electrooculography (EOG).
• Accurate classification of single-trial EEG data is essential for effective BCI performance.

Purpose of the Study:

• To develop and evaluate a novel BCI system for single-trial EEG recognition.
• To improve the accuracy of motor imagery classification by addressing EOG artifacts.
• To compare the proposed method with existing feature extraction techniques.

Main Methods:

• Independent Component Analysis (ICA) and similarity measures for automatic EOG artifact removal.
• Multiresolution asymmetry ratio for feature extraction from wavelet-transformed EEG data.
• Support Vector Machine (SVM) classifier for discriminating between motor imagery tasks (left/right finger lifting).

Main Results:

• The proposed BCI system demonstrated promising results in classifying single-trial EEG data.
• Effective removal of EOG artifacts using ICA significantly improved classification accuracy.
• The asymmetry ratio feature extraction outperformed band power and adaptive autoregressive (AAR) parameters.

Conclusions:

• The developed BCI system offers a robust approach for EEG signal recognition.
• ICA-based artifact removal is a critical step for enhancing BCI performance.
• The multiresolution asymmetry ratio is a valuable feature for motor imagery classification in BCI.