High-density force myography: A possible alternative for upper-limb prosthetic control.

Area of Science:

• Biomedical Engineering
• Rehabilitation Technology
• Prosthetics and Orthotics

Background:

• Advanced upper-limb prosthetics offer dexterous control but suffer from limited adoption due to controllability issues.
• Robust control methods are crucial for increasing the acceptance rates of prosthetic devices.

Purpose of the Study:

• To evaluate the efficacy of high-density force myography (HD-FMG) as a control method for upper-limb prosthetics.
• To compare the performance of HD-FMG against standard electromyography (EMG)-based systems for prosthetic control.

Main Methods:

• HD-FMG utilizes a high-density array of pressure sensors to detect muscle contraction patterns in the residual limb.
• Eight hand and wrist motions were classified using HD-FMG in a static arm position.
• The impact of position variation and channel reduction on classification accuracy was investigated.

Main Results:

• HD-FMG achieved a classification error of 0.33% for eight hand and wrist motions, significantly outperforming reported EMG-based methods (2.2%-11.3%).
• Incorporating position variation into the training protocol mitigated classification errors associated with limb movement.
• Informed, symmetric channel reduction further decreased classification error to an exceptional 0.02%.

Conclusions:

• HD-FMG demonstrates superior performance for prosthetic control compared to EMG, offering higher accuracy in gesture detection.
• HD-FMG is a promising technology for enhancing the controllability and acceptance of advanced upper-limb prosthetics.
• Optimized HD-FMG systems, including channel reduction, can achieve near-perfect classification accuracy for prosthetic control.