Abstract
Joint moments are critical parameters for evaluating human movement, and time-series models are widely used to predict them from biosignals. However, biosignals collected via accelerometers, gyroscopes, and electromyography (EMG) sensors are often susceptible to local features such as short-term fluctuations and noise, which hinders the models’ ability to effectively capture global features and weakens their capability to predict long-term trends. To address this issue, this paper proposes a long short-term attention memory (LSTAM) model that integrates global features. Our main contributions include the use of fast Fourier transform for spectral decomposition, multilayer perceptrons for nonlinear transformation, and convolutional modules to suppress the impact of local features in the sensor data. Additionally, an LSTM network enhanced with attention mechanisms is incorporated to dynamically focus on key temporal and frequency-domain patterns. We evaluated the proposed model on a publicly available dataset and compared its performance with existing methods, including LSTM, TCN, Conv2D, TimeMixer, xPatch, FFN, and TranSEMG. Experimental results show that the LSTAM model achieved a variance accounted for (VAF) of 0.907 ± 0.022 for hip flexion–extension (FE) and 0.927 ± 0.026 for hip abduction–adduction (AA); a root mean square error (RMSE) of 8.04 ± 2.27 (FE) and 5.56 ± 2.01 (AA); and a coefficient of determination (R2) of 0.908 ± 0.029 (FE) and 0.922 ± 0.030 (AA). These results demonstrate that LSTAM significantly outperforms existing models, offering a robust and efficient solution for joint moment prediction and human rehabilitation evaluation.
Similar content being viewed by others
Data availability
The Camargo biomechanics dataset used in this study is publicly available at http://www.epic.gatech.edu/opensource biomechanicscamargo-et-al/. The data are organized as tables following a nested directory structure of subject/date/mode/sensor.
References
Altai, Z. et al. Performance of multiple neural networks in predicting lower limb joint moments using wearable sensors. Front. Bioeng. Biotechnol. 11, 1215770 (2023).
Rabe, K. G. & Fey, N. P. Evaluating electromyography and sonomyography sensor fusion to estimate lower-limb kinematics using gaussian process regression. Front. Robot. AI 9, 716545 (2022).
Truong, M. T. N., Ali, A. E. A., Owaki, D. & Hayashibe, M. EMG-based estimation of lower limb joint angles and moments using long short-term memory network. Sensors 23, 3331 (2023).
Zhang, Q., Fragnito, N., Bao, X. & Sharma, N. A deep learning method to predict ankle joint moment during walking at different speeds with ultrasound imaging: A framework for assistive devices control. Wearable Technol. 3, e20 (2022).
Snyder, S. J. et al. Prediction of knee adduction moment using innovative instrumented insole and deep learning neural networks in healthy female individuals. Knee 41, 115–123 (2023).
Pfeiffer, M. & Hohmann, A. Applications of neural networks in training science. Hum. Mov. Sci. 31, 344–359 (2012).
Ren, L., Jones, R. K. & Howard, D. Whole body inverse dynamics over a complete gait cycle based only on measured kinematics. J. Biomech. 41, 2750–2759 (2008).
Buchanan, T. S., Lloyd, D. G., Manal, K. & Besier, T. F. Neuromusculoskeletal modeling: Estimation of muscle forces and joint moments and movements from measurements of neural command. J. Appl. Biomech. 20, 367–395 (2004).
Forner-Cordero, A., Koopman, H. & Van der Helm, F. Inverse dynamics calculations during gait with restricted ground reaction force information from pressure insoles. Gait Posture 23, 189–199 (2006).
McCabe, M. V., Van Citters, D. W. & Chapman, R. M. Developing a method for quantifying hip joint angles and moments during walking using neural networks and wearables. Comput. Methods Biomech. Biomed. Eng. 26, 1–11 (2023).
Zhong, W., Fu, X. & Zhang, M. A muscle synergy-driven ANFIS approach to predict continuous knee joint movement. IEEE Trans. Fuzzy Syst. 30, 1553–1563 (2022).
Sun, T., Li, D., Fan, B., Tan, T. & Shull, P. B. Real-time ground reaction force and knee extension moment estimation during drop landings via modular LSTM modeling and wearable IMUS. IEEE J. Biomed. Health Inform. 27, 3222–3233 (2023).
Dorschky, E. et al. CNN-based estimation of sagittal plane walking and running biomechanics from measured and simulated inertial sensor data. Front. Bioeng. Biotechnol. 8, 604 (2020).
Xiong, B. et al. Intelligent prediction of human lower extremity joint moment: An artificial neural network approach. IEEE Access 7, 29973–29980 (2019).
Scherpereel, K. L., Molinaro, D. D., Shepherd, M. K., Inan, O. T. & Young, A. J. Improving biological joint moment estimation during real-world tasks with EMG and instrumented insoles. IEEE Trans. Biomed. Eng. 71, 2718–2727 (2024).
Liu, Y.-X. & Gutierrez-Farewik, E. M. Muscle synergies enable accurate joint moment prediction using few electromyography sensors. In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 5090–5097 (IEEE, 2021).
Jamaludin, M. R. et al. Machine learning application of transcranial motor-evoked potential to predict positive functional outcomes of patients. Comput. Intell. Neurosci. 2022, 2801663 (2022).
Zhou, Y., Li, J. & Dong, M. Prediction of actively exerted torque from ankle joint complex based on muscle synergy. IEEE Trans. Ind. Electron. 71, 1729–1737. https://doi.org/10.1109/TIE.2023.3257380 (2023).
Su, C., Chen, S., Jiang, H. & Chen, Y. Ankle joint torque prediction based on surface electromyographic and angular velocity signals. IEEE Access 8, 217681–217687. https://doi.org/10.1109/ACCESS.2020.3040820 (2020).
Schimmack, M. & Mercorelli, P. Noise detection for biosignals using an orthogonal wavelet packet tree denoising algorithm. Int. J. Electron. Telecommun. 62, 15–21. https://doi.org/10.1515/ELETEL-2016-0002 (2016).
Metsis, V., Schizas, I. & Marshall, G. Real-time subspace denoising of polysomnographic data. In Proceedings of the 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments (2015). https://doi.org/10.1145/2769493.2769584.
Phinyomark, A., Nuidod, A., Phukpattaranont, P. & Limsakul, C. Feature extraction and reduction of wavelet transform coefficients for EMG pattern classification. Elektron. Elektrotech. 122, 27–32 (2012).
Englehart, K. & Hudgins, B. A robust, real-time control scheme for multifunction myoelectric control. IEEE Trans. Biomed. Eng. 50, 848–854 (2003).
Bao, T. et al. A deep Kalman filter network for hand kinematics estimation using SEMG. Pattern Recognit. Lett. 143, 88–94 (2021).
Peng, K., Guo, H. & Shang, X. EEMD and multiscale PCA-based signal denoising method and its application to seismic p-phase arrival picking. Sensors 21, 5271 (2021).
Kavran, A. J. & Clauset, A. Denoising large-scale biological data using network filters. BMC Bioinform. 22, 1–21 (2021).
Winter, D. A. Biomechanics and Motor Control of Human Movement (Wiley, Hoboken, 2009).
Xiong, B. et al. Transemg: A trans-scale hybrid model for high-accurate hip joint moment prediction. IEEE Trans. Instrum. Meas. 73, 1–11 (2024).
Sharifi-Renani, M., Mahoor, M. H. & Clary, C. W. Biomat: An open-source biomechanics multi-activity transformer for joint kinematic predictions using wearable sensors. Sensors 23, 5778 (2023).
Pan, Y. et al. Temporal re-attention LSTM for multivariate prediction in industrial heating process with large time delays. IEEE Trans. Ind. Inform. 21, 3565–3574 (2025).
Molinaro, D. D., Kang, I., Camargo, J., Gombolay, M. C. & Young, A. J. Subject-independent, biological hip moment estimation during multimodal overground ambulation using deep learning. IEEE Trans. Med. Robot. Bionics 4, 219–229 (2022).
Bao, T., Zaidi, S. A. R., Xie, S., Yang, P. & Zhang, Z.-Q. A CNN-LSTM hybrid model for wrist kinematics estimation using surface electromyography. IEEE Trans. Instrum. Meas. 70, 1–9 (2020).
Teng, H., Zhang, G., Zhou, H., Rahman, S. A. & Frisoli, A. Estimation of elbow joint angle from EMG and IMU measurements based on graph convolution neural network. In 2024 30th International Conference on Mechatronics and Machine Vision in Practice (M2VIP), 1–6 (IEEE, 2024).
Yan, J. et al. Spatial and temporal attention embedded spatial temporal graph convolutional networks for skeleton based gait recognition with multiple IMUS. Iscience 27, 110646 (2024).
Cahyadi, B. et al. Analysis of EMG based arm movement sequence using mean and median frequency. In 2018 5th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), 440–444 (IEEE, 2018).
Torres, J. F., Hadjout, D., Sebaa, A., Martínez-Álvarez, F. & Troncoso, A. Deep learning for time series forecasting: A survey. Big Data 9, 3–21 (2021).
Shepherd, M. K., Molinaro, D. D., Sawicki, G. S. & Young, A. J. Deep learning enables exoboot control to augment variable-speed walking. IEEE Robot. Autom. Lett. 7, 3571–3577 (2022).
Kang, I. et al. Real-time gait phase estimation for robotic hip exoskeleton control during multimodal locomotion. IEEE Robot. Autom. Lett. 6, 3491–3497 (2021).
Krishnan, S. & Athavale, Y. Trends in biomedical signal feature extraction. Biomed. Signal Process. Control 43, 41–63 (2018).
Sandberg, H., Mollerstedt, E. & Bernhardsson. Frequency-domain analysis of linear time-periodic systems. IEEE Trans. Autom. Control 50, 1971–1983. https://doi.org/10.1109/TAC.2005.860294 (2005).
Li, K. et al. A review of the key technologies for SEMG-based human–robot interaction systems. Biomed. Signal Process. Control 62, 102074 (2020).
Phinyomark, A., Khushaba, R. N. & Scheme, E. Feature extraction and selection for myoelectric control based on wearable EMG sensors. Sensors 18, 1615 (2018).
David Orjuela-Cañón, A., Ruíz-Olaya, A. F. & Forero, L. Deep neural network for EMG signal classification of wrist position: Preliminary results. In 2017 IEEE Latin American Conference on Computational Intelligence (LA-CCI), 1–5 (2017).
He, Y., Fukuda, O., Bu, N., Okumura, H. & Yamaguchi, N. Surface EMG pattern recognition using long short-term memory combined with multilayer perceptron. In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 5636–5639 (IEEE, 2018).
Dutra, B., Silveira, A. & Pereira, A. Grasping force estimation using state-space model and Kalman filter. Biomed. Signal Process. Control 70, 103036. https://doi.org/10.1016/j.bspc.2021.103036 (2021).
Sung, J. et al. Prediction of lower extremity multi-joint angles during overground walking by using a single IMU with a low frequency based on an LSTM recurrent neural network. Sensors 22, 53 (2021).
Vaswani, A. et al. Attention is all you need. In Advances in Neural Information Processing Systems 30 (2017).
Liu, Y.-X. & Gutierrez-Farewik, E. M. Joint kinematics, kinetics and muscle synergy patterns during transitions between locomotion modes. IEEE Trans. Biomed. Eng. 70, 1062–1071 (2022).
Camargo, J., Ramanathan, A., Flanagan, W. & Young, A. A comprehensive, open-source dataset of lower limb biomechanics in multiple conditions of stairs, ramps, and level-ground ambulation and transitions. J. Biomech. 119, 110320 (2021).
Meyer, A. J., Patten, C. & Fregly, B. J. Lower extremity EMG-driven modeling of walking with automated adjustment of musculoskeletal geometry. PLoS ONE 12, e0179698 (2017).
Kang, I., Molinaro, D. D., Choi, G., Camargo, J. & Young, A. J. Subject-independent continuous locomotion mode classification for robotic hip exoskeleton applications. IEEE Trans. Biomed. Eng. 69, 3234–3242 (2022).
Mundt, M. et al. Prediction of lower limb joint angles and moments during gait using artificial neural networks. Med. Biol. Eng. Comput. 58, 211–225 (2020).
Wang, S. et al. Timemixer: Decomposable multiscale mixing for time series forecasting. arXiv preprint arXiv:2405.14616 (2024).
Stitsyuk, A. & Choi, J. xpatch: Dual-stream time series forecasting with exponential seasonal-trend decomposition. In Proceedings of the AAAI Conference on Artificial Intelligence, vol. 39, 20601–20609 (2025).
Teoh, J. R. et al. Advancing healthcare through multimodal data fusion: A comprehensive review of techniques and applications. PeerJ Comput. Sci. 10, e2298 (2024).
Funding
This research was funded by the Regional Development Project in Fujian Province (2024H4004, 2024Y3002), the Fujian Province University-Industry Collaboration Project (2024H6016), and the Central Guidance on Local Science and Technology Development Program (2023L3030).
Author information
Authors and Affiliations
Contributions
Yinghui Guo: Performed experiments, data curation, formal analysis, writing—original draft preparation including figures, and conceptualization. Jie Lou: Data collection, writing—original draft preparation including figures, and conceptualization, and validation. Baoping Xiong: Project administration, and writing—reviewing and editing. Zhenhua Gan, Guojun Mao, Nianyin Zeng, Yong Xu: Reviewing and editing, methodological guidance and technical support.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Xiong, B., Guo, Y., Lou, J. et al. Long short-term attention memory (LSTAM): a global-feature-integrated model for joint moment prediction in human rehabilitation. Sci Rep (2026). https://doi.org/10.1038/s41598-026-42722-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-42722-6
