Name
Abstract | ||
|---|---|---|
Estimating muscle force by surface electromyography (sEMG) is a non-invasive and flexible way to diagnose biomechanical diseases and control assistive devices such as prosthetic hands. To estimate muscle force using sEMG, a supervised method is commonly adopted. This requires simultaneous recording of sEMG signals and muscle force measured by additional devices to tune the variables involved. However, recording the muscle force of the lost limb of an amputee is challenging, and the supervised method has limitations in this regard. Although the unsupervised method does not require muscle force recording, it suffers from low accuracy due to a lack of reference data. To achieve accurate and easy estimation of muscle force by the unsupervised method, we propose a decomposition of one-channel sEMG signals into constituent motor unit action potentials (MUAPs) in two steps: (1) learning an orthogonal basis of sEMG signals through reconstruction independent component analysis; (2) extracting spike-like MUAPs from the basis vectors. Nine healthy subjects were recruited to evaluate the accuracy of the proposed approach in estimating muscle force of the biceps brachii. The results demonstrated that the proposed approach based on decomposed MUAPs explains more than 80% of the muscle force variability recorded at an arbitrary force level, while the conventional amplitude-based approach explains only 62.3% of this variability. With the proposed approach, we were also able to achieve grip force control of a prosthetic hand, which is one of the most important clinical applications of the unsupervised method. Experiments on two trans-radial amputees indicated that the proposed approach improves the performance of the prosthetic hand in grasping everyday objects. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.3389/fnbot.2018.00020 | FRONTIERS IN NEUROROBOTICS |
Keywords | Field | DocType |
sEMG decomposition,reconstruction independent component analysis,motor unit action potentials,grip force estimation,prosthetic hand control | Muscle force,Biceps,Pattern recognition,Computer science,Communication channel,Electromyography,Orthogonal basis,Motor unit,Artificial intelligence,Independent component analysis,Basis (linear algebra),Machine learning | Journal |
Volume | ISSN | Citations |
12 | 1662-5218 | 1 |
PageRank | References | Authors |
0.39 | 7 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Wentao Sun | 1 | 1 | 1.40 |
| Jinying Zhu | 2 | 33 | 7.37 |
| Yinlai Jiang | 3 | 10 | 11.72 |
| Hiroshi Yokoi | 4 | 383 | 92.58 |
| Qiang Huang | 5 | 266 | 91.95 |