Functionally Adaptive Myosite Selection Using High-Density sEMG for Upper Limb Myoelectric Prostheses
IEEE Transactions on Bio-Medical Engineering
|May 16, 2023
Summary
Functionally Adaptive Myosite Selection (FAMS) is a new method for placing electrodes in prosthetic limbs. It quickly predicts performance, improving control and reducing fitting time for amputees.
Area of Science:
- Biomedical Engineering
- Rehabilitation Engineering
- Prosthetics
Background:
- Effective prosthetic control relies on optimal electrode placement.
- Current methods can be time-consuming and may not adapt to individual patient needs.
Purpose of the Study:
- To introduce Functionally Adaptive Myosite Selection (FAMS), a novel method for rapid and effective electrode placement in prosthesis fitting.
- To demonstrate FAMS's adaptability to patient anatomy and functional goals, independent of classification models.
- To provide insight into expected classifier performance without extensive model training.
Main Methods:
- FAMS utilizes a separability metric to predict classifier performance during prosthesis fitting.
- The method evaluates electrode configurations based on predicted performance and anatomical suitability.
- Applied to two amputee subjects to determine optimal electrode placement.
Main Results:
- A predictable relationship exists between the FAMS metric and classifier accuracy, enabling performance estimation.
- FAMS-selected configurations improved control performance with Artificial Neural Network (ANN) classifiers and matched Linear Discriminant Analysis (LDA) classifier performance.
- Achieved 95.8% of maximum classification performance using a mean of 25 electrodes for amputee subjects.
Conclusions:
- FAMS offers a rapid tool for approximating the trade-offs between electrode count and classifier performance.
- This method is valuable for optimizing electrode placement during prosthesis fitting.
- FAMS enhances the efficiency and effectiveness of prosthetic limb control for amputees.
