Real-time decoding of brain responses to visuospatial attention using 7T fMRI

Area of Science:

• Neuroscience
• Biomedical Engineering
• Medical Imaging

Background:

• Brain-Computer interfaces (BCIs) offer communication for severe paralysis.
• Invasive BCI recordings offer improved signal quality and novel control strategies.
• Pre-surgical fMRI is crucial for determining optimal invasive electrode implantation sites.

Purpose of the Study:

• To investigate eye movement-independent visuospatial attention as a BCI control strategy.
• To assess the feasibility of using real-time 7T fMRI for detecting localized brain activity for BCI control.
• To evaluate the trainability of visuospatial attention for BCI applications.

Main Methods:

• Real-time functional Magnetic Resonance Imaging (fMRI) at 7 Tesla was employed.
• Healthy subjects focused attention on spatial targets while maintaining central gaze.
• Electrographic Coregistration (ECoG) data was analyzed for gamma-power changes.

Main Results:

• Real-time fMRI successfully detected subtle, localized brain activity changes related to visuospatial attention.
• All subjects could regulate brain activity in targeted regions without prior training.
• ECoG data indicated that gamma-power changes could differentiate attentional states.

Conclusions:

• Visuospatial attention is a viable and trainable control strategy for invasive BCIs.
• High-field real-time fMRI is effective for pre-surgical planning and BCI development.
• This approach holds potential for enhancing communication and independence in individuals with paralysis.