A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation
Simultaneous implementation of myoelectric pattern recognition and muscle force estimation is highly demanded in building natural gestural interfaces but a challenging task due to the gesture classification accuracy degradation under varying muscle strengths. To address this problem, a novel method...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10194336/ |
| _version_ | 1797741035475959808 |
|---|---|
| author | Xinhui Li Xu Zhang Liwei Zhang Xiang Chen Ping Zhou |
| author_facet | Xinhui Li Xu Zhang Liwei Zhang Xiang Chen Ping Zhou |
| author_sort | Xinhui Li |
| collection | DOAJ |
| description | Simultaneous implementation of myoelectric pattern recognition and muscle force estimation is highly demanded in building natural gestural interfaces but a challenging task due to the gesture classification accuracy degradation under varying muscle strengths. To address this problem, a novel method using transformer-based multi-task learning (MTL-Transformer) for the prediction of both myoelectric patterns and corresponding muscle strengths was proposed to describe the inherent characteristics of an individual gesture pattern under different force conditions, thereby improving the accuracy of myoelectric pattern recognition. In addition, the transformer model enabled the characterization of long-term temporal correlations to ensure precise and smooth estimation of the muscle force. The performance of the proposed MTL-Transformer framework was evaluated via experiments of classifying eleven hand gestures and estimating the corresponding muscle force simultaneously, using high-density surface electromyogram (HD-sEMG) recordings from forearm flexor muscles of eleven intact-limbed subjects. The MTL-Transformer framework yielded high classification accuracy (98.70±1.21%) and low root mean square deviation (12.59±2.76%), and outperformed other two common temporally modelling methods significantly (<inline-formula> <tex-math notation="LaTeX">${p} < 0.05$ </tex-math></inline-formula>) in terms of both improved gesture recognition accuracies and reduced muscle force estimation errors. The MTL-Transformer framework is demonstrated as an effective solution for simultaneous implementation of myoelectric pattern recognition and muscle force estimation. This study promotes the development of robust and smooth myoelectric control systems, with wide applications in gestural interfaces, prosthetic and orthotic control. |
| first_indexed | 2024-03-12T14:21:04Z |
| format | Article |
| id | doaj.art-066329209b4c4d82a020d1b1a8014c6a |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-12T14:21:04Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-066329209b4c4d82a020d1b1a8014c6a2023-08-18T23:00:07ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-01313255326410.1109/TNSRE.2023.329879710194336A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force EstimationXinhui Li0Xu Zhang1https://orcid.org/0000-0002-1533-4340Liwei Zhang2Xiang Chen3https://orcid.org/0000-0001-8259-4815Ping Zhou4https://orcid.org/0000-0002-4394-2677School of Microelectronics, University of Science and Technology of China, Hefei, ChinaSchool of Microelectronics, University of Science and Technology of China, Hefei, ChinaFirst Affiliated Hospital, University of Science and Technology of China, Hefei, ChinaSchool of Microelectronics, University of Science and Technology of China, Hefei, ChinaDepartment of Biomedical and Rehabilitation Engineering, University of Health and Rehabilitation Sciences, Qingdao, ChinaSimultaneous implementation of myoelectric pattern recognition and muscle force estimation is highly demanded in building natural gestural interfaces but a challenging task due to the gesture classification accuracy degradation under varying muscle strengths. To address this problem, a novel method using transformer-based multi-task learning (MTL-Transformer) for the prediction of both myoelectric patterns and corresponding muscle strengths was proposed to describe the inherent characteristics of an individual gesture pattern under different force conditions, thereby improving the accuracy of myoelectric pattern recognition. In addition, the transformer model enabled the characterization of long-term temporal correlations to ensure precise and smooth estimation of the muscle force. The performance of the proposed MTL-Transformer framework was evaluated via experiments of classifying eleven hand gestures and estimating the corresponding muscle force simultaneously, using high-density surface electromyogram (HD-sEMG) recordings from forearm flexor muscles of eleven intact-limbed subjects. The MTL-Transformer framework yielded high classification accuracy (98.70±1.21%) and low root mean square deviation (12.59±2.76%), and outperformed other two common temporally modelling methods significantly (<inline-formula> <tex-math notation="LaTeX">${p} < 0.05$ </tex-math></inline-formula>) in terms of both improved gesture recognition accuracies and reduced muscle force estimation errors. The MTL-Transformer framework is demonstrated as an effective solution for simultaneous implementation of myoelectric pattern recognition and muscle force estimation. This study promotes the development of robust and smooth myoelectric control systems, with wide applications in gestural interfaces, prosthetic and orthotic control.https://ieeexplore.ieee.org/document/10194336/Myoelectric pattern recognitionmuscle force estimationvarying muscle strengthstransformer modelmulti-task learning |
| spellingShingle | Xinhui Li Xu Zhang Liwei Zhang Xiang Chen Ping Zhou A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation IEEE Transactions on Neural Systems and Rehabilitation Engineering Myoelectric pattern recognition muscle force estimation varying muscle strengths transformer model multi-task learning |
| title | A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation |
| title_full | A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation |
| title_fullStr | A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation |
| title_full_unstemmed | A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation |
| title_short | A Transformer-Based Multi-Task Learning Framework for Myoelectric Pattern Recognition Supporting Muscle Force Estimation |
| title_sort | transformer based multi task learning framework for myoelectric pattern recognition supporting muscle force estimation |
| topic | Myoelectric pattern recognition muscle force estimation varying muscle strengths transformer model multi-task learning |
| url | https://ieeexplore.ieee.org/document/10194336/ |
| work_keys_str_mv | AT xinhuili atransformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT xuzhang atransformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT liweizhang atransformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT xiangchen atransformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT pingzhou atransformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT xinhuili transformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT xuzhang transformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT liweizhang transformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT xiangchen transformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation AT pingzhou transformerbasedmultitasklearningframeworkformyoelectricpatternrecognitionsupportingmuscleforceestimation |