Exploring Kinematics Information Decoding from EEG Slow Cortical Potentials During Movement Imagination and...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Motor Imagery-based brain-computer interfaces (MI-BCIs) are advancing for enhanced interaction and assistive technologies.
• Decoding kinematic information from EEG slow cortical potentials (SCPs) is crucial for MI-BCI development.

Purpose of the Study:

• To investigate the feasibility of decoding hand movement direction and speed from EEG SCPs during motor imagery.
• To analyze the reliability of SCP features for decoding kinematic information.

Main Methods:

• EEG data from 14 healthy subjects imagining hand movements at different directions and speeds were analyzed.
• Movement-related cortical potential peak negativity from 15 motor cortex channels was used for decoding.
• Pearson correlation was applied for subject-specific channel selection.

Main Results:

• Average accuracy for decoding direction-speed pairs was 63.44% for motor imagery.
• Speed classification accuracy was around 53.87%, not significantly different from chance.
• Direction decoding showed higher accuracy than speed decoding in both imagined and observed movements.

Conclusions:

• EEG SCPs contain reliable directional information for motor imagery BCIs.
• Decoding movement speed from SCPs is less consistent and requires further research.
• Findings suggest potential for SCP-based directional control and open avenues for studying movement observation.