Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks
Joint torque prediction is crucial when investigating biomechanics, evaluating treatments, and designing powered assistive devices. Controllers in assistive technology require reference torque trajectories to set the level of assistance for a patient during rehabilitation or when aiding essential da...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute of Electrical and Electronics Engineers
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/114882/ http://psasir.upm.edu.my/id/eprint/114882/1/114882.pdf |
| _version_ | 1848866623717376000 |
|---|---|
| author | Perera, Chamalka Kenneth Gopalai, Alpha A. Gouwanda, Darwin Ahmad, Siti A. Teh, Pei-Lee |
| author_facet | Perera, Chamalka Kenneth Gopalai, Alpha A. Gouwanda, Darwin Ahmad, Siti A. Teh, Pei-Lee |
| author_sort | Perera, Chamalka Kenneth |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Joint torque prediction is crucial when investigating biomechanics, evaluating treatments, and designing powered assistive devices. Controllers in assistive technology require reference torque trajectories to set the level of assistance for a patient during rehabilitation or when aiding essential daily tasks such as sit-to-walk (STW). STW itself can be generalized into strategies based on individual needs and movement patterns. In this study, three long short-term memory (LSTM) neural networks were empirically trained for hip and knee torque prediction considering these STW strategies and subject anthropometry. The hip and knee are the drivers of STW, while the network architectures were selected for recognizing temporal and spatial relationships. Performance of the LSTMs were compared and evaluated against the STW strategies to accurately generate strategy-specific and user-oriented torque. As such, train and test STW data were obtained from 65 subjects across three age groups: young, middle-aged, and older adults (19-73 years). Model inputs were hip and knee angles with horizontal center of mass velocity, while windowing allowed the LSTMs to dynamically adapt to real-time changes in STW transitions. The encoder-decoder LSTM showcased optimal performance with robust recognition of temporal features. It produced significantly (P< 0.05) low hip and knee root mean square error (0.24±0.07 and 0.15±0.02 Nm/kg), strong Spearman's correlation (93.43±2.86 and 84.83±2.96%) and good intraclass correlation coefficients (greater than 0.75), demonstrating model reliability. Hence, this network predicts strategy and user oriented reference torques for personalized controllers in assistive devices, with more natural application of assistance. |
| first_indexed | 2025-11-15T14:23:33Z |
| format | Article |
| id | upm-114882 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:23:33Z |
| publishDate | 2024 |
| publisher | Institute of Electrical and Electronics Engineers |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1148822025-02-06T08:11:26Z http://psasir.upm.edu.my/id/eprint/114882/ Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks Perera, Chamalka Kenneth Gopalai, Alpha A. Gouwanda, Darwin Ahmad, Siti A. Teh, Pei-Lee Joint torque prediction is crucial when investigating biomechanics, evaluating treatments, and designing powered assistive devices. Controllers in assistive technology require reference torque trajectories to set the level of assistance for a patient during rehabilitation or when aiding essential daily tasks such as sit-to-walk (STW). STW itself can be generalized into strategies based on individual needs and movement patterns. In this study, three long short-term memory (LSTM) neural networks were empirically trained for hip and knee torque prediction considering these STW strategies and subject anthropometry. The hip and knee are the drivers of STW, while the network architectures were selected for recognizing temporal and spatial relationships. Performance of the LSTMs were compared and evaluated against the STW strategies to accurately generate strategy-specific and user-oriented torque. As such, train and test STW data were obtained from 65 subjects across three age groups: young, middle-aged, and older adults (19-73 years). Model inputs were hip and knee angles with horizontal center of mass velocity, while windowing allowed the LSTMs to dynamically adapt to real-time changes in STW transitions. The encoder-decoder LSTM showcased optimal performance with robust recognition of temporal features. It produced significantly (P< 0.05) low hip and knee root mean square error (0.24±0.07 and 0.15±0.02 Nm/kg), strong Spearman's correlation (93.43±2.86 and 84.83±2.96%) and good intraclass correlation coefficients (greater than 0.75), demonstrating model reliability. Hence, this network predicts strategy and user oriented reference torques for personalized controllers in assistive devices, with more natural application of assistance. Institute of Electrical and Electronics Engineers 2024 Article PeerReviewed text en cc_by_4 http://psasir.upm.edu.my/id/eprint/114882/1/114882.pdf Perera, Chamalka Kenneth and Gopalai, Alpha A. and Gouwanda, Darwin and Ahmad, Siti A. and Teh, Pei-Lee (2024) Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 32. pp. 3977-3986. ISSN 1534-4320; eISSN: 1558-0210 https://ieeexplore.ieee.org/document/10739358/ 10.1109/TNSRE.2024.3488052 |
| spellingShingle | Perera, Chamalka Kenneth Gopalai, Alpha A. Gouwanda, Darwin Ahmad, Siti A. Teh, Pei-Lee Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title | Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title_full | Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title_fullStr | Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title_full_unstemmed | Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title_short | Lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| title_sort | lower limb torque prediction for sit-to-walk strategies using long short-term memory neural networks |
| url | http://psasir.upm.edu.my/id/eprint/114882/ http://psasir.upm.edu.my/id/eprint/114882/ http://psasir.upm.edu.my/id/eprint/114882/ http://psasir.upm.edu.my/id/eprint/114882/1/114882.pdf |