Decoding of four movement directions using hybrid NIRS-EEG brain-computer interface.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Brain-computer interfaces (BCIs) offer novel ways to control external devices using brain activity.
• Hybrid BCIs combining multiple neuroimaging modalities can improve signal acquisition and classification accuracy.
• Existing BCI systems face challenges in achieving robust and precise control for complex tasks.

Purpose of the Study:

• To develop and evaluate a hybrid near-infrared spectroscopy-electroencephalography (NIRS-EEG) BCI system.
• To decode four distinct brain signals for directional control commands (forward, backward, left, right).
• To assess the classification accuracy of the hybrid NIRS-EEG approach.

Main Methods:

• An experimental hybrid NIRS-EEG system was implemented.
• NIRS sensors were placed over the prefrontal cortex, and EEG sensors over the motor cortex.
• Twelve participants were presented with directional cues, performing mental arithmetic for forward/backward and hand tapping for left/right commands.

Main Results:

• The hybrid NIRS-EEG technique successfully extracted and decoded four different brain signal types.
• High classification accuracies were achieved for all four directional control signals.
• Oxy-hemoglobin changes (via NIRS) correlated with mental arithmetic tasks, while motor cortex activity (via EEG) correlated with hand tapping.

Conclusions:

• The hybrid NIRS-EEG BCI technology demonstrates high potential for accurate brain-signal classification.
• This multimodal approach enables precise formulation of control commands for BCI applications.
• The findings support the efficacy of combining NIRS and EEG for advanced BCI development.