Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue
This study proposes a novel method for monitoring muscle fatigue using muscle-specific dynamic models which relate joint time-frequency signatures extracted from the relevant electromyogram (EMG) signals with the corresponding estimated muscle forces. Muscle forces were estimated using physics-drive...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Prognostics and Health Management Society
2022-07-01
|
| Series: | International Journal of Prognostics and Health Management |
| Subjects: | |
| Online Access: | https://papers.phmsociety.org/index.php/ijphm/article/view/3132 |
| _version_ | 1828419166641586176 |
|---|---|
| author | Haihua Ou Deanna Gates Shane Johnson Dragan Djurdjanovic |
| author_facet | Haihua Ou Deanna Gates Shane Johnson Dragan Djurdjanovic |
| author_sort | Haihua Ou |
| collection | DOAJ |
| description | This study proposes a novel method for monitoring muscle fatigue using muscle-specific dynamic models which relate joint time-frequency signatures extracted from the relevant electromyogram (EMG) signals with the corresponding estimated muscle forces. Muscle forces were estimated using physics-driven musculoskeletal models which incorporate muscle lengths and contraction velocities estimated from the available kinematic and kinetic measurements. For any specific individual, such a muscle-specific dynamic model is trained using EMG and movement data collected in the early stages of an exercise, i.e., during the least-fatigued behavior. As the exercise or physical activity of that individual progresses and fatigue develops, residuals yielded by that model when approximating the newly arrived data shift and change because of the fatigue-induced changes in the underlying dynamics. In this paper, we propose quantitative evaluation of those changes via the concept of a muscle-specific Freshness Index (FI) which at any given time expresses overlaps between the distribution of that muscle’s model residuals observed on the most recently collected data and the distribution of modeling residuals observed during non-fatigued behavior. The newly proposed method was evaluated using data collected during a repetitive sawing motion experiment with 12 healthy participants. The performance of the FI as a fatigue metric was compared with the performance of the instantaneous frequency of the relevant EMG signals, which is a more traditional and widely used metric of muscle fatigue. It was found that the FI reflected the progression of muscle fatigue with desirable properties of stronger monotonic trends and smaller noise levels compared to the traditional, instantaneous frequency-based metrics. |
| first_indexed | 2024-12-10T14:48:02Z |
| format | Article |
| id | doaj.art-64c54bfeed124ee2baf42d6c27ee3ee3 |
| institution | Directory Open Access Journal |
| issn | 2153-2648 |
| language | English |
| last_indexed | 2024-12-10T14:48:02Z |
| publishDate | 2022-07-01 |
| publisher | The Prognostics and Health Management Society |
| record_format | Article |
| series | International Journal of Prognostics and Health Management |
| spelling | doaj.art-64c54bfeed124ee2baf42d6c27ee3ee32022-12-22T01:44:30ZengThe Prognostics and Health Management SocietyInternational Journal of Prognostics and Health Management2153-26482022-07-01Vol. 13No.2https://doi.org/10.36001/ijphm.2022.v13i2.3132Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle FatigueHaihua Ou0Deanna Gates1 Shane Johnson2Dragan Djurdjanovic3UM-SJTU Joint Institute, Shanghai Jiao Tong University, Shanghai, Shanghai, 200240, ChinaSchool of Kinesiology, University of Michigan, Ann Arbor, MI48109, USAUM-SJTU Joint Institute, Shanghai Jiao Tong University, Shanghai, Shanghai, 200240, ChinaDepartment of Mechanical Engineering, University of Texas, Austin, Texas 78712, USAThis study proposes a novel method for monitoring muscle fatigue using muscle-specific dynamic models which relate joint time-frequency signatures extracted from the relevant electromyogram (EMG) signals with the corresponding estimated muscle forces. Muscle forces were estimated using physics-driven musculoskeletal models which incorporate muscle lengths and contraction velocities estimated from the available kinematic and kinetic measurements. For any specific individual, such a muscle-specific dynamic model is trained using EMG and movement data collected in the early stages of an exercise, i.e., during the least-fatigued behavior. As the exercise or physical activity of that individual progresses and fatigue develops, residuals yielded by that model when approximating the newly arrived data shift and change because of the fatigue-induced changes in the underlying dynamics. In this paper, we propose quantitative evaluation of those changes via the concept of a muscle-specific Freshness Index (FI) which at any given time expresses overlaps between the distribution of that muscle’s model residuals observed on the most recently collected data and the distribution of modeling residuals observed during non-fatigued behavior. The newly proposed method was evaluated using data collected during a repetitive sawing motion experiment with 12 healthy participants. The performance of the FI as a fatigue metric was compared with the performance of the instantaneous frequency of the relevant EMG signals, which is a more traditional and widely used metric of muscle fatigue. It was found that the FI reflected the progression of muscle fatigue with desirable properties of stronger monotonic trends and smaller noise levels compared to the traditional, instantaneous frequency-based metrics.https://papers.phmsociety.org/index.php/ijphm/article/view/3132muscle fatiguesystem-based performance monitoringemg signalstime-frequency signal analysis |
| spellingShingle | Haihua Ou Deanna Gates Shane Johnson Dragan Djurdjanovic Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue International Journal of Prognostics and Health Management muscle fatigue system-based performance monitoring emg signals time-frequency signal analysis |
| title | Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue |
| title_full | Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue |
| title_fullStr | Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue |
| title_full_unstemmed | Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue |
| title_short | Model-based Fusion of Surface Electromyography with Kinematic and Kinetic Measurements for Monitoring of Muscle Fatigue |
| title_sort | model based fusion of surface electromyography with kinematic and kinetic measurements for monitoring of muscle fatigue |
| topic | muscle fatigue system-based performance monitoring emg signals time-frequency signal analysis |
| url | https://papers.phmsociety.org/index.php/ijphm/article/view/3132 |
| work_keys_str_mv | AT haihuaou modelbasedfusionofsurfaceelectromyographywithkinematicandkineticmeasurementsformonitoringofmusclefatigue AT deannagates modelbasedfusionofsurfaceelectromyographywithkinematicandkineticmeasurementsformonitoringofmusclefatigue AT shanejohnson modelbasedfusionofsurfaceelectromyographywithkinematicandkineticmeasurementsformonitoringofmusclefatigue AT dragandjurdjanovic modelbasedfusionofsurfaceelectromyographywithkinematicandkineticmeasurementsformonitoringofmusclefatigue |