Brain computer interface: estimation of cortical activity from non invasive high resolution EEG recordings

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Non-invasive electroencephalography (EEG) is a key tool for monitoring brain activity.
• Detecting mental states like motor imagery is crucial for brain-computer interfaces (BCIs).
• Current EEG methods may have limitations in precisely localizing cortical activity related to imagined movements.

Purpose of the Study:

• To assess the utility of estimated cortical activity from non-invasive EEG for detecting mental states associated with limb movement imagination.
• To compare the efficacy of estimated cortical activity versus surface EEG recordings for motor imagery detection.

Main Methods:

• High-resolution EEG data from five healthy subjects imagining limb movements were analyzed.
• Realistic head models were used to estimate cortical activity.
• Depth-weighted minimum norm solutions were applied to Brodmann areas of interest.
• Comparisons were made between surface EEG and estimated cortical activity.

Main Results:

• Estimated cortical activity during limb motor imagery was primarily localized to the contralateral primary motor area.
• A greater imbalance in brain activity between contralateral and ipsilateral motor areas was observed with estimated cortical activity compared to surface EEG.
• This statistically significant unbalance was more pronounced in the estimated cortical activity.

Conclusions:

• Estimated cortical activity shows potential superiority over surface EEG for detecting upper limb motor imagery.
• The enhanced detection capability stems from a greater statistically significant hemispheric unbalance.
• Findings support the advancement of non-invasive brain-computer interfaces using estimated cortical activity.