Phantom-Mobility-Based Prosthesis Control in Transhumeral Amputees Without Surgical Reinnervation: A Preliminary...

Area of Science:

• Biomedical Engineering
• Neuroscience
• Rehabilitation Technology

Background:

• Transhumeral amputees experience challenges with current myoelectric prosthetics, leading to high rejection rates.
• Existing control methods are often slow and unnatural, particularly for prostheses with multiple degrees of freedom.
• Surgical muscle reinnervation offers enhanced control but requires invasive procedures.

Purpose of the Study:

• To investigate the feasibility of using voluntary phantom limb mobilization (PLM) for controlling advanced prosthetic arms.
• To assess if PLM-based myoelectric pattern recognition enables natural and functional grasping tasks for transhumeral amputees.
• To evaluate this non-surgical approach as an alternative to muscle reinnervation.

Main Methods:

• Two transhumeral amputees performed repetitive grasping and releasing tasks with an eight-active-DoF prosthetic arm.
• Prosthesis control relied on myoelectric pattern recognition trained using minimal PLM data (two repetitions).
• Performance was compared against a control group using a graphical user interface (GUI).

Main Results:

• Participants successfully completed functional grasping tasks using PLM-based control.
• Optimal strategies and joint trajectories were observed, though task completion time increased.
• Control strategies required frequent corrections, indicating areas for improvement.

Conclusions:

• PLM-based myoelectric control is a potentially viable, non-surgical option for transhumeral amputees.
• Further research is needed to address robustness of pattern recognition and real-world prosthesis perturbations.
• This approach could enhance control of multi-joint prosthetic arms, offering an alternative to muscle reinnervation surgery.