MEG-based neurofeedback for hand rehabilitation

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Science

Background:

• Brain-computer interfaces (BCI) can enhance therapeutic plasticity through intuitive neurofeedback (NF).
• Developing a BCI for direct control of motor-related brain activity is key for maximizing BCI utility.
• This study presents a novel BCI for intuitive control of a video-based grasp.

Purpose of the Study:

• To develop and evaluate a magnetoencephalography (MEG)-based BCI system.
• To provide realistic and intuitive neurofeedback (NF) for motor control.
• To assess the potential of this BCI for inducing neuroplasticity in individuals with paralysis.

Main Methods:

• Magnetoencephalography (MEG) recorded sensorimotor rhythms (SMR) modulated by grasp/rest intentions.
• SMR controlled the aperture of a virtual hand in a stop-motion video.
• Task difficulty was adjusted by changing the required hold time for targeted grasp positions.

Main Results:

• Three individuals with spinal cord injury (SCI)-induced paralysis achieved 63% average success in controlling a virtual hand.
• Performance significantly exceeded the chance rate (19%) without extensive pre-training or calibration.
• Two participants demonstrated learning, evidenced by significant SMR improvement within a single NF session.

Conclusions:

• A MEG-based BCI system effectively delivers focused neurofeedback (NF) for motor control in paralysis.
• The system demonstrated utility in providing realistic and efficient NF for inducing neuroplasticity.
• This approach holds potential for rehabilitation in individuals with paralysis.