Related Experiment Video
Updated: Aug 19, 2025
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
700
Intramuscular EMG feature extraction and evaluation at different arm positions and hand postures based on a
Ali Asghar1,2, Saad Jawaid Khan1, Fahad Azim2
1Department of Biomedical Engineering, Faculty of Engineering, Science, Technology and Management, Ziauddin University, Karachi, Pakistan.
Summary
This study identifies optimal electromyography (EMG) signal features for controlling prosthetic devices. Variance (VAR) excels in fixed arm positions, while Average Amplitude Change (AAC) is best for fixed hand postures, enhancing prosthetic control.
Area of Science:
- Biomedical Engineering
- Rehabilitation Sciences
- Assistive Technology
Background:
- Electromyography (EMG) signals are crucial for controlling prostheses and assistive devices.
- Pattern recognition and machine learning enhance the classification of limb motions for prosthetic applications.
Purpose of the Study:
- To propose a feature extraction and evaluation method for intramuscular electromyography (iEMG) signals.
- To identify optimal time-domain features for prosthetic control under varying arm positions and hand postures using the RES Index statistical criterion.
Main Methods:
- Utilized sixteen time-domain features from iEMG signals of eight healthy males performing five motion classes.
- Collected data at four arm positions (0°, 45°, 90°, 135°) under fixed arm position (FAP) and fixed hand posture (FHP) conditions.
- Employed k-nearest neighbor (KNN) classifier and calculated the RES Index to evaluate feature performance.
Main Results:
- For fixed arm positions (FAP), Variance (VAR) was the best feature, while WAMP, Zero Crossing (ZC), and Slope Sign Change (SSC) were least effective.
- For fixed hand postures (FHP), Average Amplitude Change (AAC) was the optimal feature, with WAMP and Simple Square Integral (SSI) performing poorly.
Conclusions:
- The study identified distinct optimal iEMG features for prosthetic control based on arm position and hand posture.
- Findings support the development of advanced iEMG-based assistive control devices and robotics.
- Future work should explore frequency-domain features for enhanced prosthetic control.
