Related Experiment Video
Updated: Sep 16, 2025
06:58
A Structured Rehabilitation Protocol for Improved Multifunctional Prosthetic Control: A Case Study
Published on: November 6, 2015
9.6K
Stimulation in Training Data to Improve Proportional Motor Control in a Dexterous Sensorimotor Task
Summary
Electrotactile stimulation for prosthetic sensory feedback can corrupt surface electromyography (sEMG) signals. However, incorporating this artifact during training may improve control accuracy for bidirectional prostheses.
Area of Science:
- Biomedical Engineering
- Neuroscience
- Rehabilitation Engineering
Background:
- Electrocautaneous stimulation offers a cost-effective method for sensory feedback in prosthetics.
- Signal artifacts from electrocutaneous stimulation can interfere with surface electromyography (sEMG) signals, potentially leading to motor control errors.
Purpose of the Study:
- To investigate the impact of electrocutaneous stimulation artifact on sEMG-based control of a virtual hand.
- To determine if incorporating stimulation artifact during training improves motor control performance.
Main Methods:
- Eight volunteers controlled a 5-degree-of-freedom virtual hand using sEMG.
- Sensory feedback was provided via electrocutaneous stimulation proportional to index finger force.
- Training data with and without random stimulation artifact was compared for its effect on discriminating object compliance.
Main Results:
- While not statistically significant, random stimulation during training improved discrimination of object compliance from 76% to 82% and 72% to 85% for two control algorithms.
- The tested algorithms included a modified Kalman filter and a convolutional neural network.
Conclusions:
- Incorporating electrocutaneous stimulation artifact into the training data shows potential for enhancing motor control in bidirectional prostheses.
- This method requires no additional computational resources and offers improved performance for prosthetic control systems.
