Related Experiment Video
Updated: Mar 9, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
1.4K
A motion-classification strategy based on sEMG-EEG signal combination for upper-limb amputees
Xiangxin Li1,2, Oluwarotimi Williams Samuel1,2, Xu Zhang1,3
1Chinese Academy of Sciences (CAS) Key Laboratory of Human-Machine Intelligence-Synergy Systems, Shenzhen Institutes of Advanced Technology, Shenzhen, 518055, China.
Journal of Neuroengineering and Rehabilitation
|January 8, 2017
Summary
Combining surface electromyography (sEMG) and electroencephalography (EEG) signals significantly improves prosthetic control for above-elbow amputees. This fusion enhances motion classification accuracy, paving the way for more functional myoelectric prostheses.
Area of Science:
- Biomedical Engineering
- Neuroscience
- Rehabilitation Technology
Background:
- Modern prostheses often rely on surface electromyography (sEMG) from residual limb muscles.
- Limited residual muscles, especially after above-elbow amputations, restrict control for multi-degree-of-freedom prostheses.
- Signal fusion, combining sEMG with non-EMG signals, offers a solution for insufficient control commands.
Purpose of the Study:
- To investigate a motion-classification method combining sEMG and electroencephalography (EEG) signals.
- To enhance the control performance of upper-limb prostheses for amputees.
- To improve motion intention decoding by integrating diverse biosignals.
Main Methods:
- Four transhumeral amputees participated in experiments.
- Simultaneous acquisition of sEMG and EEG signals during five specified motion classes.
- Independent signal preprocessing, feature extraction (time-domain), and classification using Linear Discriminant Analysis (LDA).
- Sequential Forward Selection (SFS) algorithm applied for channel optimization.
Main Results:
- Fusion of sEMG and EEG yielded significantly better classification performance than individual signal sources.
- A combination of 32-channel sEMG and 64-channel EEG increased classification accuracy by over 14%.
- Optimized electrode arrangements (10-channel sEMG + 10-channel EEG, 10-channel sEMG + 20-channel EEG) achieved 84.2% and 87.0% accuracy, respectively.
Conclusions:
- Feasibility of fusing sEMG and EEG signals demonstrated for improving motion classification accuracy in above-elbow amputees.
- This approach holds potential for enhancing the clinical control performance of multifunctional myoelectric prostheses.
- Signal fusion offers a promising strategy to overcome limitations of single-signal-based prosthetic control.
