Training-Efficient Personalised Knee Joint Moment Estimation via an Incremental Broad Learning System

Guoyu Zuo , Minghui Zhang , Jingwei Ren , Qifei Wu , Jiaxin Li , Jianyu Chen , Chengyu Qiao , Shuangyue Yu

CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (4) : 1179 -1193.

PDF (3885KB)
CAAI Transactions on Intelligence Technology ›› 2026, Vol. 11 ›› Issue (4) :1179 -1193. DOI: 10.1049/cit2.70165
ORIGINAL RESEARCH
research-article
Training-Efficient Personalised Knee Joint Moment Estimation via an Incremental Broad Learning System
Author information +
History +
PDF (3885KB)

Abstract

Knee joint moment estimation is a critical component in biomechanical analysis, with profound implications for rehabilitation assessment and the development of assistive devices like exoskeletons. However, existing data-driven approaches often rely on highly complex network architectures in pursuit of high accuracy, resulting in inherently inefficient training procedures. Furthermore, incorporating new subject-specific data typically requires retraining the entire model, which substantially increases computational cost. These limitations make it challenging for current data-driven methods to achieve personalised knee joint moment estimation. To address this challenge, this paper proposes a novel framework based on an Incremental Broad Learning System (IBLS) that can achieve efficient model training through a flattened network architecture and an analytical ridge regression solver. More importantly, it integrates an incremental learning algorithm that enables rapid model updates using only newly acquired subject-specific data, eliminating the need for costly full retraining. These capabilities collectively make efficient personalised knee joint moment estimation feasible. Extensive experiments on Dataset A and Dataset B demonstrate that our method achieves higher prediction accuracy compared to standard baselines (ANN and LSTM). Furthermore, our approach exhibits competitive performance when compared with advanced deep learning architectures (Transformer and TCN-LSTM). Notably, the training efficiency is significantly enhanced: the training time required by our method is approximately 30% of that for ANN, 7.5% for LSTM, and only 3.5% for both the Transformer and TCN-LSTM. Further experimental evidence indicates that incorporating incremental learning provides an additional 60% reduction in training time relative to full retraining, without sacrificing prediction accuracy. In addition, we systematically identify optimal Inertial Measurement Unit (IMU) configurations that balance accuracy with practical wearability, providing actionable guidelines for implementation. This work provides an efficient, accurate, and personalised solution for knee joint moment estimation, paving the way for the development of adaptive and efficient wearable robotic systems.

Keywords

incremental broad learning system / joint torque estimation / knee joint moment / wearable robots

Cite this article

Download citation ▾
Guoyu Zuo, Minghui Zhang, Jingwei Ren, Qifei Wu, Jiaxin Li, Jianyu Chen, Chengyu Qiao, Shuangyue Yu. Training-Efficient Personalised Knee Joint Moment Estimation via an Incremental Broad Learning System. CAAI Transactions on Intelligence Technology, 2026, 11 (4) : 1179-1193 DOI:10.1049/cit2.70165

登录浏览全文

4963

注册一个新账户 忘记密码

References

[1]

W. L. Kenney, J. H. Wilmore, and D. L. Costill, Physiology of Sport and Exercise (Human Kinetics, 2021).

[2]

C. Wang, P. P. Chan, B. M. Lam, et al., “ Real-Time Estimation of Knee Adduction Moment for Gait Retraining in Patients With Knee Osteoarthritis,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 28, no. 4 (2020): 888-894, https://doi.org/10.1109/tnsre.2020.2978537.

[3]

J.-H. Kim, S.-M. Lee, H.-S. Shin, et al., “ Biomechanical Mechanisms Behind the Reduction of Knee Adduction Moment in Medial Knee Thrust Gait,” Scientific Reports 15, no. 1 (2025): 37317, https://doi.org/10.1038/s41598-025-21220-1.

[4]

N. D. Reeves and F. L. Bowling, “ Conservative Biomechanical Strategies for Knee Osteoarthritis,” Nature Reviews Rheumatology 7, no. 2 (2011): 113-122, https://doi.org/10.1038/nrrheum.2010.212.

[5]

B. Jackson, A. Wluka, A. Teichtahl, M. E. Morris, and F. M. Cicuttini, “ Reviewing Knee Osteoarthritis-A Biomechanical Perspective,” Journal of Science and Medicine in Sport 7, no. 3 (2004): 347-357, https://doi.org/10.1016/s1440-2440(04)80030-6.

[6]

D. D. Molinaro, I. Kang, and A. J. Young, “ Estimating Human Joint Moments Unifies Exoskeleton Control, Reducing User Effort,” Science Robotics 9, no. 88 (2024): eadi8852, https://doi.org/10.1126/scirobotics.adi8852.

[7]

G. M. Gasparri, J. Luque, and Z. F. Lerner, “ Proportional Joint Moment Control for Instantaneously Adaptive Ankle Exoskeleton Assistance,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 27, no. 4 (2019): 751-759, https://doi.org/10.1109/tnsre.2019.2905979.

[8]

T.-H. Huang, S. Zhang, S. Yu, et al., “ Modeling and Stiffness-Based Continuous Torque Control of Lightweight Quasi-Direct-Drive Knee Exoskeletons for Versatile Walking Assistance,” IEEE Transactions on Robotics 38, no. 3 (2022): 1442-1459, https://doi.org/10.1109/tro.2022.3170287.

[9]

X. Wu, A. Zhu, X. Wang, X. Ma, H. Mao, and J. Song, “ Real-Time Gait Symmetry Enhancement in People With Unilateral Knee Injuries Using Deep Learning for Modulation of Knee Exoskeleton,” in 2023 International Conference on Design Science (ICDS) (IEEE, 2023), 1-7.

[10]

W. Meng, Z. Tian, C. Zhu, Q. Ai, and Q. Liu, “ Optimized Impedance Control of a Lightweight Gait Rehabilitation Exoskeleton Based on Accurate Knee Joint Torque Estimation,” IEEE Transactions on Medical Robotics and Bionics 6, no. 4 (2024): 1648-1657, https://doi.org/10.1109/tmrb.2024.3464671.

[11]

J. Zhang, P. Fiers, K. A. Witte, et al., “ Human-in-the-Loop Optimization of Exoskeleton Assistance During Walking,” Science 356, no. 6344 (2017): 1280-1284, https://doi.org/10.1126/science.aal5054.

[12]

Y. Ding, M. Kim, S. Kuindersma, and C. J. Walsh, “ Human-in-the-Loop Optimization of Hip Assistance With a Soft Exosuit During Walking,” Science Robotics 3, no. 15 (2018): eaar5438, https://doi.org/10.1126/scirobotics.aar5438.

[13]

M. S. B. Hossain, Z. Guo, and H. Choi, “ Estimation of Lower Extremity Joint Moments and 3d Ground Reaction Forces Using Imu Sensors in Multiple Walking Conditions: A Deep Learning Approach,” IEEE Journal of Biomedical and Health Informatics 27, no. 6 (2023): 2829-2840, https://doi.org/10.1109/jbhi.2023.3262164.

[14]

H. Lim, B. Kim, and S. Park, “ Prediction of Lower Limb Kinetics and Kinematics During Walking by a Single IMU on the Lower Back Using Machine Learning,” Sensors 20, no. 1 (2019): 130, https://doi.org/10.3390/s20010130.

[15]

K. L. Scherpereel, D. D. Molinaro, M. K. Shepherd, O. T. Inan, and A. J. Young, “ Improving Biological Joint Moment Estimation During Real-World Tasks With EMG and Instrumented Insoles,” IEEE Transactions on Biomedical Engineering 71, no. 9 (2024): 2718-2727, https://doi.org/10.1109/tbme.2024.3388874.

[16]

L. Zhang, Z. Li, Y. Hu, C. Smith, E. M. G. Farewik, and R. Wang, “ Ankle Joint Torque Estimation Using an EMG-Driven Neuromusculoskeletal Model and an Artificial Neural Network Model,” IEEE Transactions on Automation Science and Engineering 18, no. 2 (2020): 564-573, https://doi.org/10.1109/tase.2020.3033664.

[17]

N. Hondo and T. Tsuji, “ Torque Estimation of Knee Flexion and Extension Movements From a Mechanomyogram of the Femoral Muscle,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 30 (2022): 1120-1126, https://doi.org/10.1109/tnsre.2022.3169225.

[18]

L. Zhang, D. Soselia, R. Wang, and E. M. Gutierrez-Farewik, “ Estimation of Joint Torque by EMG-Driven Neuromusculoskeletal Models and LSTM Networks,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 31 (2023): 3722-3731, https://doi.org/10.1109/tnsre.2023.3315373.

[19]

L. Zhang, D. Soselia, R. Wang, and E. M. Gutierrez-Farewik, “ Lower-Limb Joint Torque Prediction Using LSTM Neural Networks and Transfer Learning,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 30 (2022): 600-609, https://doi.org/10.1109/tnsre.2022.3156786.

[20]

W. Niu, Q. Bian, L. Yang, H. Zhou, and Z. Ding, “ Cross-LSTM: Integrating Cross Attention With Long Short-Term Memory Neural Networks in Estimating Joint Moments From Wearable Sensors,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 33 (2025): 3793-3804, https://doi.org/10.1109/TNSRE.2025.3610281.

[21]

H. J. Bennett, K. Estler, K. Valenzuela, and J. T. Weinhandl, “ Predicting Knee Joint Contact Forces During Normal Walking Using Kinematic Inputs With a Long-Short Term Neural Network,” Journal of Biomechanical Engineering 146, no. 8 (2024): 081004, https://doi.org/10.1115/1.4064550.

[22]

J. Zhang, Y. Zhao, F. Shone, et al., “ Physics-Informed Deep Learning for Musculoskeletal Modeling: Predicting Muscle Forces and Joint Kinematics From Surface EMG,” IEEE Transactions on Neural Systems and Rehabilitation Engineering 31 (2022): 484-493, https://doi.org/10.1109/tnsre.2022.3226860.

[23]

Z. Wang, C. Chen, H. Chen, Y. Zhou, X. Wang, and X. Wu, “ Dual Transformer Network for Predicting Joint Angles and Torques From Multi-Channel EMG Signals in the Lower Limbs,” IEEE Journal of Biomedical and Health Informatics (2025), https://doi.org/10.1109/JBHI.2025.3555255.

[24]

B. Xiong, Y. Chang, N. Zeng, et al., “ TranSEMG: A Trans-Scale Hybrid Model for High-Accurate Hip Joint Moment Prediction,” IEEE Transactions on Instrumentation and Measurement 73 (2024): 1-11, https://doi.org/10.1109/tim.2024.3384565.

[25]

Z. Feng, D. Zhu, T. Wu, H. Li, and T. Han, “Self-Aware Fusion IMU-EMG Attention Dependence for Knee Adduction Moment Estimation During Walking,” IEEE Journal of Biomedical and Health Informatics 6, no. 4 (2025): 1648-1657, https://doi.org/10.1109/JBHI.2025.3564981.

[26]

D. D. Molinaro, K. L. Scherpereel, E. B. Schonhaut, G. Evangelopoulos, M. K. Shepherd, and A. J. Young, “ Task-Agnostic Exoskeleton Control via Biological Joint Moment Estimation,” Nature 635, no. 8038 (2024): 337-344, https://doi.org/10.1038/s41586-024-08157-7.

[27]

D. D. Molinaro, I. Kang, J. Camargo, M. C. Gombolay, and A. J. Young, “ Subject-Independent, Biological Hip Moment Estimation During Multimodal Overground Ambulation Using Deep Learning,” IEEE Transactions on Medical Robotics and Bionics 4, no. 1 (2022): 219-229, https://doi.org/10.1109/tmrb.2022.3144025.

[28]

C. P. Chen, Z. Liu, and S. Feng, “ Universal Approximation Capability of Broad Learning System and its Structural Variations,” IEEE Transactions on Neural Networks and Learning Systems 30, no. 4 (2018): 1191-1204, https://doi.org/10.1109/tnnls.2018.2866622.

[29]

C. P. Chen and Z. Liu, “ Broad Learning System: An Effective and Efficient Incremental Learning System Without the Need for Deep Architecture,” IEEE Transactions on Neural Networks and Learning Systems 29, no. 1 (2017): 10-24, https://doi.org/10.1109/tnnls.2017.2716952.

[30]

Z. Liu, C. P. Chen, S. Feng, Q. Feng, and T. Zhang, “ Stacked Broad Learning System: From Incremental Flatted Structure to Deep Model,” IEEE Transactions on Systems, Man, and Cybernetics: Systems 51, no. 1 (2020): 209-222, https://doi.org/10.1109/tsmc.2020.3043147.

[31]

L. Zhang, J. Li, G. Lu, et al., “ Analysis and Variants of Broad Learning System,” IEEE Transactions on Systems, Man, and Cybernetics: Systems 52, no. 1 (2020): 334-344, https://doi.org/10.1109/tsmc.2020.2995205.

[32]

H. Huang, T. Zhang, C. Yang, and C. P. Chen, “ Motor Learning and Generalization Using Broad Learning Adaptive Neural Control,” IEEE Transactions on Industrial Electronics 67, no. 10 (2019): 8608-8617, https://doi.org/10.1109/tie.2019.2950853.

[33]

S. Xu, J. Liu, C. Yang, X. Wu, and T. Xu, “ A Learning-Based Stable Servo Control Strategy Using Broad Learning System Applied for Microrobotic Control,” IEEE Transactions on Cybernetics 52, no. 12 (2021): 13727-13737, https://doi.org/10.1109/tcyb.2021.3121080.

[34]

S. Xu, T. Xu, D. Li, C. Yang, C. Huang, and X. Wu, “ A Robot Motion Learning Method Using Broad Learning System Verified by Small-Scale Fish-Like Robot,” IEEE Transactions on Cybernetics 53, no. 9 (2023): 6053-6065, https://doi.org/10.1109/tcyb.2023.3269773.

[35]

Y. Cao, M. Jia, P. Ding, X. Zhao, and Y. Ding, “ Incremental Learning for Remaining Useful Life Prediction via Temporal Cascade Broad Learning System With Newly Acquired Data,” IEEE Transactions on Industrial Informatics 19, no. 4 (2023): 6234-6245, https://doi.org/10.1109/tii.2022.3201977.

[36]

T. Xu, G. Han, H. Zhu, C. Lin, and J. Peng, “ Multiscale BLS-Based Lightweight Prediction Model for Remaining Useful Life of Aero-Engine,” IEEE Transactions on Reliability 73, no. 4 (2024): 1757-1767, https://doi.org/10.1109/tr.2023.3349201.

[37]

C. Zhu, Y. Pu, Z. Lyu, K. Yang, and Q. Yang, “ Multidimensional Information Fusion and Broad Learning System-Based Condition Recognition for Energy Pipeline Safety,” Knowledge-Based Systems 300 (2024): 112259, https://doi.org/10.1016/j.knosys.2024.112259.

[38]

X. Xu, S. Bao, H. Shao, and P. Shi, “ A Multi Sensor Fused Incremental Broad Learning With DS Theory for Online Fault Diagnosis of Rotating Machinery,” Advanced Engineering Informatics 60 (2024): 102419, https://doi.org/10.1016/j.aei.2024.102419.

[39]

J. Camargo, A. Ramanathan, W. Flanagan, and A. Young, “ A Comprehensive, Open-Source Dataset of Lower Limb Biomechanics in Multiple Conditions of Stairs, Ramps, and Level-Ground Ambulation and Transitions,” Journal of Biomechanics 119 (2021): 110320, https://doi.org/10.1016/j.jbiomech.2021.110320.

[40]

K. Scherpereel, D. Molinaro, O. Inan, M. Shepherd, and A. Young, “ A Human Lower-Limb Biomechanics and Wearable Sensors Dataset During Cyclic and Non-Cyclic Activities,” Scientific Data 10, no. 1 (2023): 924, https://doi.org/10.1038/s41597-023-02840-6.

[41]

A. Seth, M. Sherman, J. A. Reinbolt, and S. L. Delp, “ OpenSim: A Musculoskeletal Modeling and Simulation Framework for in Silico Investigations and Exchange,” Procedia IUTAM 2 (2011): 212-232, https://doi.org/10.1016/j.piutam.2011.04.021.

[42]

Z. Tang, Y. Wang, S. Mao, and D. Wei, “ TCN-xLSTM: A Hybrid Temporal Model Integrating TCN and xLSTM for Lower Limb Joint Moment Estimation From IMU Signals,” IEEE Robotics and Automation Letters 11, no. 1 (2025): 153-160, https://doi.org/10.1109/lra.2025.3632757.

[43]

B. Hur, S. Baek, I. Kang, and D. Kim, “Learning Based Lower Limb Joint Kinematic Estimation Using Open Source IMU Data,” Scientific Reports 15, no. 1 (2025): 5287, https://doi.org/10.1038/s41598-025-89716-4.

PDF (3885KB)

0

Accesses

0

Citation

Detail

Sections
Recommended

/

〈 〉