A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation
Seated postural limit defines the boundary of a region such that for any excursions made outside this boundary a subject cannot return the trunk to the neutral position without additional external support. The seated postural limits can be used as a reference to provide assistive support to the tors...
| 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/9944704/ |
| _version_ | 1797805072360407040 |
|---|---|
| author | Xupeng Ai Victor Santamaria Jiawei Chen Boce Hu Chenfei Zhu Sunil K. Agrawal |
| author_facet | Xupeng Ai Victor Santamaria Jiawei Chen Boce Hu Chenfei Zhu Sunil K. Agrawal |
| author_sort | Xupeng Ai |
| collection | DOAJ |
| description | Seated postural limit defines the boundary of a region such that for any excursions made outside this boundary a subject cannot return the trunk to the neutral position without additional external support. The seated postural limits can be used as a reference to provide assistive support to the torso by the Trunk Support Trainer (TruST). However, fixed boundary representations of seated postural limits are inadequate to capture dynamically changing seated postural limits during training. In this study, we propose a conceptual model of dynamic boundary of the trunk center by assigning a vector that tracks the postural-goal direction and trunk movement amplitude during a sitting task. We experimented with 20 healthy subjects. The results support our hypothesis that TruST intervention with an assist-as-needed force controller based on dynamic boundary representation could achieve more significant sitting postural control improvements than a fixed boundary representation. The second contribution of this paper is that we provide an effective approach to embed deep learning into TruST’s real-time controller design. We have compiled a 3D trunk movement dataset which is currently the largest in the literature. We designed a loss function capable of solving the gate-controlled regression problem. We have proposed a novel deep-learning roadmap for the exploration study. Following the roadmap, we developed a deep learning architecture, modified the widely used Inception module, and then obtained a deep learning model capable of accurately predicting the dynamic boundary in real-time. We believe that this approach can be extended to other rehabilitation robots towards designing intelligent dynamic boundary-based assist-as-needed controllers. |
| first_indexed | 2024-03-13T05:46:26Z |
| format | Article |
| id | doaj.art-2bf21495cfdf4b81b21c98458b9149c7 |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-13T05:46:26Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-2bf21495cfdf4b81b21c98458b9149c72023-06-13T20:09:26ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102023-01-013126027010.1109/TNSRE.2022.32213089944704A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk RehabilitationXupeng Ai0https://orcid.org/0000-0003-4155-9348Victor Santamaria1https://orcid.org/0000-0002-7889-6957Jiawei Chen2Boce Hu3Chenfei Zhu4Sunil K. Agrawal5https://orcid.org/0000-0002-4008-1437Department of Mechanical Engineering, Columbia University, New York, NY, USADepartment of Mechanical Engineering, Columbia University, New York, NY, USADepartment of Mechanical Engineering, Columbia University, New York, NY, USADepartment of Mechanical Engineering, Columbia University, New York, NY, USADepartment of Mechanical Engineering, Columbia University, New York, NY, USADepartment of Mechanical Engineering, and the Department of Rehabilitation and Regenerative Medicine, Columbia University, New York, NY, USASeated postural limit defines the boundary of a region such that for any excursions made outside this boundary a subject cannot return the trunk to the neutral position without additional external support. The seated postural limits can be used as a reference to provide assistive support to the torso by the Trunk Support Trainer (TruST). However, fixed boundary representations of seated postural limits are inadequate to capture dynamically changing seated postural limits during training. In this study, we propose a conceptual model of dynamic boundary of the trunk center by assigning a vector that tracks the postural-goal direction and trunk movement amplitude during a sitting task. We experimented with 20 healthy subjects. The results support our hypothesis that TruST intervention with an assist-as-needed force controller based on dynamic boundary representation could achieve more significant sitting postural control improvements than a fixed boundary representation. The second contribution of this paper is that we provide an effective approach to embed deep learning into TruST’s real-time controller design. We have compiled a 3D trunk movement dataset which is currently the largest in the literature. We designed a loss function capable of solving the gate-controlled regression problem. We have proposed a novel deep-learning roadmap for the exploration study. Following the roadmap, we developed a deep learning architecture, modified the widely used Inception module, and then obtained a deep learning model capable of accurately predicting the dynamic boundary in real-time. We believe that this approach can be extended to other rehabilitation robots towards designing intelligent dynamic boundary-based assist-as-needed controllers.https://ieeexplore.ieee.org/document/9944704/Seated postural limitsassist-as-needed controller (AANC)rehabilitation roboticssupervised learningstatistical machine learning |
| spellingShingle | Xupeng Ai Victor Santamaria Jiawei Chen Boce Hu Chenfei Zhu Sunil K. Agrawal A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation IEEE Transactions on Neural Systems and Rehabilitation Engineering Seated postural limits assist-as-needed controller (AANC) rehabilitation robotics supervised learning statistical machine learning |
| title | A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation |
| title_full | A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation |
| title_fullStr | A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation |
| title_full_unstemmed | A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation |
| title_short | A Deep-Learning Based Real-Time Prediction of Seated Postural Limits and Its Application in Trunk Rehabilitation |
| title_sort | deep learning based real time prediction of seated postural limits and its application in trunk rehabilitation |
| topic | Seated postural limits assist-as-needed controller (AANC) rehabilitation robotics supervised learning statistical machine learning |
| url | https://ieeexplore.ieee.org/document/9944704/ |
| work_keys_str_mv | AT xupengai adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT victorsantamaria adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT jiaweichen adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT bocehu adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT chenfeizhu adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT sunilkagrawal adeeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT xupengai deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT victorsantamaria deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT jiaweichen deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT bocehu deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT chenfeizhu deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation AT sunilkagrawal deeplearningbasedrealtimepredictionofseatedposturallimitsanditsapplicationintrunkrehabilitation |