A hybrid brain-muscle-machine interface for stroke rehabilitation: Usability and functionality validation in a 2-week...

Area of Science:

• Neuroscience
• Rehabilitation Engineering
• Biomedical Engineering

Background:

• Non-invasive brain-machine interfaces (BMIs) face limitations in spatial resolution for stroke rehabilitation.
• Hybrid brain-muscle-machine interfaces (hBMIs) offer enhanced control by integrating muscle signals.
• Controlling multi-degree-of-freedom (DoF) exoskeletons with hBMIs presents challenges in distributing control between brain and muscle.

Purpose of the Study:

• To evaluate the feasibility, usability, and functionality of a bio-inspired hBMI for controlling a 7-DoF upper limb exoskeleton.
• To assess the potential of hBMI in facilitating motor relearning and improving function in chronic stroke patients.
• To investigate the hierarchical control strategy and mirror myoelectric decoder for enhanced rehabilitation.

Main Methods:

• A hierarchical control strategy mimicking natural motor pathways was implemented.
• An innovative mirror myoelectric decoder was used to guide patients in relearning muscle activation patterns.
• Six chronic, severely paralyzed stroke patients participated in a 2-week hBMI training program with a 7-DoF exoskeleton, followed by physiotherapy.

Main Results:

• Patients reported high adoption rates and confidence in the hBMI technology.
• Significant improvements were observed in arm function (50% of patients) and hand function (83% of patients).
• Enhanced muscle activation patterns and increased motor evoked potentials were noted post-intervention.

Conclusions:

• Bio-inspired hBMIs show considerable potential for rehabilitating severely paralyzed chronic stroke patients.
• Integrating brain and muscle signals via hBMIs can effectively improve motor control and functional recovery.
• The developed hBMI system facilitates coordinated movements, aiding in the early reintegration of affected limbs into daily activities.