A Gait Imagery-Based Brain-Computer Interface With Visual Feedback for Spinal Cord Injury Rehabilitation on Lokomat.

Area of Science:

• Neuroscience
• Rehabilitation Engineering
• Biomedical Engineering

Background:

• Spinal Cord Injury (SCI) presents significant challenges for motor function restoration.
• Motor Imagery (MI)-based Brain-Computer Interfaces (BCIs) offer a promising avenue for neurorehabilitation.
• Integrating BCIs with assistive devices like the Lokomat platform can enhance therapeutic interventions.

Purpose of the Study:

• To develop and evaluate an Electroencephalography (EEG) based BCI system utilizing gait imagery for motor recovery in SCI individuals.
• To investigate the efficacy of a novel Visual Neurofeedback (VNFB) system in modulating cortical rhythms associated with gait.
• To explore the simultaneous activation of central and peripheral nervous mechanisms for enhanced rehabilitation.

Main Methods:

• A multi-channel EEG system was employed to discriminate gait imagery tasks during walking.
• A Visual Neurofeedback (VNFB) system was implemented, targeting specific brain rhythms (8-12 Hz and 15-20 Hz) around the Cz electrode.
• A cluster analysis strategy based on Euclidean distance was used, with a weighted mean MI feature vector as a reference for user training.

Main Results:

• The developed BCI system achieved an average classification accuracy of 74.4%.
• Feature analysis revealed a reduction in cluster variance over several sessions, indicating improved user control.
• Metrics demonstrated increased separation between classes (passive walking with and without gait MI), suggesting effective self-modulation of cortical rhythms.

Conclusions:

• Intervention using a gait MI-based BCI with VNFB shows potential for enabling individuals with SCI to modulate their target cortical rhythms.
• This approach contributes to the advancement of gait rehabilitation systems by integrating Machine Learning and neurofeedback.
• The study highlights the potential for restoring lower-limb functions in SCI individuals through innovative BCI technology.