Related Experiment Video
Updated: Jun 28, 2026
06:44
Development of a Novel Task-oriented Rehabilitation Program using a Bimanual Exoskeleton Robotic Hand
Published on: May 20, 2020
Use of RPNIs and Implanted Electrodes for Prosthetic Wrist and Multi-Grip Hand Control During Functional Tasks: A
IEEE Transactions on Bio-Medical Engineering
|June 26, 2026
Summary
Implanted intramuscular EMG signals from regenerative peripheral nerve interfaces (RPNIs) improved prosthetic control and function compared to surface EMG. This approach offers better performance and lower cognitive load for amputees using myoelectric prostheses.
Area of Science:
- Biomedical Engineering
- Neuroprosthetics
- Rehabilitation Science
Background:
- Surface electromyography (sEMG) pattern recognition for prosthetic control is limited by electrode shift and poor signal quality.
- Intramuscular electrodes and regenerative peripheral nerve interfaces (RPNIs) offer improved signal fidelity and additional control channels.
- Current systems struggle with reliable control of multiple degrees of freedom, such as wrist rotation.
Purpose of the Study:
- To compare functional, biomechanical, and cognitive outcomes of prosthetic control using sEMG versus implanted intramuscular EMG (imEMG) from RPNIs and residual muscles.
- To evaluate the benefits of active wrist rotation control in a myoelectric prosthesis.
- To explore the potential advantages of combining EMG signal sources for enhanced prosthetic control.
Main Methods:
- A case study involving one female participant with a unilateral transradial amputation.
- Five control approaches were tested using a myoelectric prosthesis with pattern recognition classifiers.
- Classifiers decoded sEMG and/or imEMG from residual muscles and RPNIs for functional grips with or without wrist rotation.
Main Results:
- Participant performance on the Clothespin Relocation Test and Coffee Task improved with imEMG compared to sEMG.
- Active wrist rotation provided modest reduction in trunk compensation (2-7°).
- imEMG with wrist rotation resulted in significantly lower cognitive workload compared to sEMG (58 pts reduction), with a combined classifier yielding zero cognitive workload.
Conclusions:
- EMG signals from implanted electrodes in RPNIs enabled better functional performance and control of multiple grips and active wrist rotation compared to sEMG, without increasing cognitive burden.
- Implanted imEMG signals from RPNIs and residual muscles show significant potential for enhancing daily prosthetic control.
- This approach could lead to more intuitive and reliable control of upper-limb prostheses.
