Enhanced Motor Imagery Decoding by Calibration Model-Assisted With Tactile ERD

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Brain-Computer Interfaces (BCIs) enable communication and control through brain activity.
• Motor Imagery (MI) is a common BCI paradigm requiring user calibration.
• Current MI calibration methods can be time-consuming and may lack optimal accuracy.

Purpose of the Study:

• To introduce and evaluate a novel tactile stimulation-assisted calibration method for MI-based BCIs.
• To compare the performance and efficiency of the proposed method against conventional calibration techniques.
• To investigate the neural correlates underlying the tactile-assisted calibration process.

Main Methods:

• A tactile stimulation-assisted motor imagery (SA-MI) task was developed, applying tactile input to the wrist.
• Classifier training was performed using data from the SA-MI task and conventional MI task.
• System performance was evaluated on a standard MI dataset, comparing SA-MI Calibration with Conventional Calibration.

Main Results:

• SA-MI Calibration achieved significantly higher decoding accuracy (78.3%) compared to Conventional Calibration (71.3%).
• The average calibration time was reduced by 40% using the SA-MI Calibration method.
• SA-MI induced greater motor-related cortical activation and higher R² values in the alpha-beta band compared to MI.

Conclusions:

• The proposed SA-MI Calibration method offers a substantial improvement in BCI performance and efficiency.
• Tactile stimulation during calibration enhances neural activity relevant to MI tasks.
• This method presents a promising approach for faster and more accurate BCI system setup.