Federated LSTM-LRP Networks with SMOTE and LIME for Interpretable Gait Classification
Received: 8 September 2025 | Revised: 23 September 2025, 17 October 2025, 25 October 2025, and 2 November 2025 | Accepted: 3 November 2025 | Online: 8 December 2025
Corresponding author: O. Pushpalatha
Abstract
Gait analysis is essential for the diagnosis and monitoring of various biomechanical and neurological conditions. Clinical reliability is compromised due to imbalanced data and lack of transparency in existing models. This study uses vertical GRF features of patients to analyze gait on the publicly available GaitRec dataset. The proposed approach addresses the limitations of traditional models using Borderline SMOTE to balance the dataset and FedAvg-SVRG to stabilize gradient updates across dispersed nodes and enable federated learning with stable convergence without centralizing data. Enhanced LRPs are combined with LSTM to improve transparency in pattern recognition, and LIME offers instance-level interpretability. A deep learning model is used to improve the accuracy of gait classification, achieving 99.72% on single run execution and 99.45±0.14% for 10-fold cross-validation. The proposed deep learning framework holds significant potential for robust gait classification and clinical decision making.
Keywords:
federated stochastic variance reduced gradient, interpretable gait analysis network, Local Interpretable Model-agnostic Explanations (LIME)Downloads
References
S. P. Lalehzarian, A. K. Gowd, and J. N. Liu, "Machine learning in orthopaedic surgery," World Journal of Orthopedics, vol. 12, no. 9, pp. 685–699, Sep. 2021. DOI: https://doi.org/10.5312/wjo.v12.i9.685
R. Das, S. Paul, G. K. Mourya, N. Kumar, and M. Hussain, "Recent Trends and Practices Toward Assessment and Rehabilitation of Neurodegenerative Disorders: Insights From Human Gait," Frontiers in Neuroscience, vol. 16, Apr. 2022. DOI: https://doi.org/10.3389/fnins.2022.859298
Y. Matsushita, D. T. Tran, H. Yamazoe, and J. H. Lee, "Recent use of deep learning techniques in clinical applications based on gait: a survey," Journal of Computational Design and Engineering, vol. 8, no. 6, pp. 1499–1532, Oct. 2021. DOI: https://doi.org/10.1093/jcde/qwab054
R. Griffiths et al., "Guideline for the management of hip fractures 2020," Anaesthesia, vol. 76, no. 2, pp. 225–237, 2021. DOI: https://doi.org/10.1111/anae.15291
B. Su, C. Smith, and E. Gutierrez Farewik, "Gait Phase Recognition Using Deep Convolutional Neural Network with Inertial Measurement Units," Biosensors, vol. 10, no. 9, Sep. 2020, Art. no. 109. DOI: https://doi.org/10.3390/bios10090109
A. S. Alharthi, S. U. Yunas, and K. B. Ozanyan, "Deep learning for monitoring of human gait: A review," IEEE Sensors Journal, vol. 19, no. 21, pp. 9575–9591, 2019. DOI: https://doi.org/10.1109/JSEN.2019.2928777
J. Cuadrado, F. Michaud, U. Lugrís, and M. Pérez Soto, "Using Accelerometer Data to Tune the Parameters of an Extended Kalman Filter for Optical Motion Capture: Preliminary Application to Gait Analysis," Sensors, vol. 21, no. 2, Jan. 2021, Art. no. 427. DOI: https://doi.org/10.3390/s21020427
I. K. Jalata, T. D. Truong, J. L. Allen, H. S. Seo, and K. Luu, "Movement Analysis for Neurological and Musculoskeletal Disorders Using Graph Convolutional Neural Network," Future Internet, vol. 13, no. 8, Aug. 2021, Art. no. 194. DOI: https://doi.org/10.3390/fi13080194
J. Kubicek, F. Tomanec, M. Cerny, D. Vilimek, M. Kalova, and D. Oczka, "Recent Trends, Technical Concepts and Components of Computer-Assisted Orthopedic Surgery Systems: A Comprehensive Review," Sensors, vol. 19, no. 23, Jan. 2019, Art. no. 5199. DOI: https://doi.org/10.3390/s19235199
N. Kour, S. Gupta, and S. Arora, "A Survey of Knee Osteoarthritis Assessment Based on Gait," Archives of Computational Methods in Engineering, vol. 28, no. 2, pp. 345–385, Mar. 2021. DOI: https://doi.org/10.1007/s11831-019-09379-z
F. Horst, S. Lapuschkin, W. Samek, K. R. Müller, and W. I. Schöllhorn, "Explaining the unique nature of individual gait patterns with deep learning," Scientific Reports, vol. 9, no. 1, Feb. 2019, Art. no. 2391. DOI: https://doi.org/10.1038/s41598-019-38748-8
Z. Zhang et al., "Deep learning-enabled triboelectric smart socks for IoT-based gait analysis and VR applications," npj Flexible Electronics, vol. 4, no. 1, Oct. 2020, Art. no. 29. DOI: https://doi.org/10.1038/s41528-020-00092-7
Y. Zhang and Y. Ma, "Application of supervised machine learning algorithms in the classification of sagittal gait patterns of cerebral palsy children with spastic diplegia," Computers in Biology and Medicine, vol. 106, pp. 33–39, Mar. 2019. DOI: https://doi.org/10.1016/j.compbiomed.2019.01.009
M. Hnatiuc, O. Geman, A. G. Avram, D. Gupta, and K. Shankar, "Human Signature Identification Using IoT Technology and Gait Recognition," Electronics, vol. 10, no. 7, Jan. 2021, Art. no. 852. DOI: https://doi.org/10.3390/electronics10070852
E. Warmerdam, M. Orth, T. Pohlemann, and B. Ganse, "Gait Analysis to Monitor Fracture Healing of the Lower Leg," Bioengineering, vol. 10, no. 2, Feb. 2023, Art. no. 255. DOI: https://doi.org/10.3390/bioengineering10020255
A. Fändriks, R. Tranberg, J. Karlsson, M. Möller, and R. Zügner, "Gait biomechanics in patients with intra-articular tibial plateau fractures – gait analysis at three months compared with age- and gender-matched healthy subjects," BMC Musculoskeletal Disorders, vol. 22, no. 1, Aug. 2021, Art. no. 702. DOI: https://doi.org/10.1186/s12891-021-04577-y
Y. L. Ng, X. Jiang, Y. Zhang, S. B. Shin, and R. Ning, "Automated Activity Recognition with Gait Positions Using Machine Learning Algorithms," Engineering, Technology & Applied Science Research, vol. 9, no. 4, pp. 4554–4560, Aug. 2019. DOI: https://doi.org/10.48084/etasr.2952
A. I. Bulbul, U. Mayetin, and S. Kucuk, "Development of an Ankle Sensor for Ground Reaction Force Measurement in Intelligent Prosthesis," Engineering, Technology & Applied Science Research, vol. 14, no. 4, pp. 15161–15170, Aug. 2024. DOI: https://doi.org/10.48084/etasr.7430
J. Perazzone, S. Wang, M. Ji, and K. S. Chan, "Communication-efficient device scheduling for federated learning using stochastic optimization," in IEEE INFOCOM 2022-IEEE Conference on Computer Communications, 2022, pp. 1449–1458. DOI: https://doi.org/10.1109/INFOCOM48880.2022.9796818
R. Johnson and T. Zhang, "Accelerating Stochastic Gradient Descent using Predictive Variance Reduction," in Advances in Neural Information Processing Systems, 2013, vol. 26.
R. Xin, S. Kar, and U. A. Khan, "Decentralized stochastic optimization and machine learning: A unified variance-reduction framework for robust performance and fast convergence," IEEE Signal Processing Magazine, vol. 37, no. 3, pp. 102–113, 2020. DOI: https://doi.org/10.1109/MSP.2020.2974267
P. J. Werbos, "Backpropagation through time: what it does and how to do it," Proceedings of the IEEE, vol. 78, no. 10, pp. 1550–1560, 1990. DOI: https://doi.org/10.1109/5.58337
S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural computation, vol. 9, no. 8, pp. 1735–1780, 1997. DOI: https://doi.org/10.1162/neco.1997.9.8.1735
A. Binder, S. Bach, G. Montavon, K. R. Müller, and W. Samek, "Layer-Wise Relevance Propagation for Deep Neural Network Architectures," in Information Science and Applications (ICISA) 2016, vol. 376, K. J. Kim and N. Joukov, Eds. Springer Singapore, 2016, pp. 913–922. DOI: https://doi.org/10.1007/978-981-10-0557-2_87
L. Arras et al., "Explaining and Interpreting LSTMs," in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, vol. 11700, W. Samek, G. Montavon, A. Vedaldi, L. K. Hansen, and K. R. Müller, Eds. Springer International Publishing, 2019, pp. 211–238. DOI: https://doi.org/10.1007/978-3-030-28954-6_11
B. Horsak, Djordje Slijepcevic, A. M. Raberger, C. Schwab, M. Worisch, and M. Zeppelzauer, "GaitRec: A large-scale ground reaction force dataset of healthy and impaired gait." figshare, 2020.
B. Horsak, D. Slijepcevic, A. M. Raberger, C. Schwab, M. Worisch, and M. Zeppelzauer, "GaitRec, a large-scale ground reaction force dataset of healthy and impaired gait," Scientific Data, vol. 7, no. 1, May 2020, Art. no. 143. DOI: https://doi.org/10.1038/s41597-020-0481-z
D. Jani et al., "An Efficient Gait Abnormality Detection Method Based on Classification," Journal of Sensor and Actuator Networks, vol. 11, no. 3, Jun. 2022, Art. no. 31. DOI: https://doi.org/10.3390/jsan11030031
M. M. Mulwa, R. W. Mwangi, and A. Mindila, "GMM‐LIME explainable machine learning model for interpreting sensor‐based human gait," Engineering Reports, vol. 6, no. 10, Oct. 2024, Art. no. e12864. DOI: https://doi.org/10.1002/eng2.12864
M. Savadkoohi, T. Oladunni, and L. A. Thompson, "Deep neural networks for human’s fall-risk prediction using force-plate time series signal," Expert Systems with Applications, vol. 182, Nov. 2021, Art. no. 115220. DOI: https://doi.org/10.1016/j.eswa.2021.115220
S. A. Boompelli and S. Bhattacharya, "Design of a telemetric gait analysis insole and 1-D convolutional neural network to track postoperative fracture rehabilitation," in 2021 IEEE 3rd Global Conference on Life Sciences and Technologies (LifeTech), 2021, pp. 484–488. DOI: https://doi.org/10.1109/LifeTech52111.2021.9391975
M. Iber et al., "Mind the Steps: Towards Auditory Feedback in Tele-Rehabilitation Based on Automated Gait Classification," in Audio Mostly 2021, Sep. 2021, pp. 139–146. DOI: https://doi.org/10.1145/3478384.3478398
C. Pandey et al., "GaitRec-Net: A Deep Neural Network for Gait Disorder Detection Using Ground Reaction Force," PPAR Research, vol. 2022, pp. 1–10, Aug. 2022. DOI: https://doi.org/10.1155/2022/9355015
Downloads
How to Cite
License
Copyright (c) 2025 O. Pushpalatha, R. Premkumar
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
