P. S. Saritha; S. Poonguzhali; B. Mohan

Adaptive Myoelectric Prosthetic Control Using Hybrid CNN–Vision Transformer–LSTM Networks and Federated Reinforcement Learning

PDF (997KB), PP.272-297

Views: 0 Downloads: 0

Author(s)

P. S. Saritha ¹ S. Poonguzhali ² B. Mohan ^3,*

1. Department of Electronics, Sri Ramakrishna College of Arts and Science, Coimbatore, India

2. Department of Physics, Sri Ramakrishna College of Arts and Science, Coimbatore, India

3. Department of ECE, P.S.R. Engineering College, Sivakasi, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2026.05.15

Received: 17 Jun. 2026 / Revised: 3 Jul. 2026 / Accepted: 14 Sep. 2026 / Published: 8 Oct. 2026

Index Terms

Smart Prosthetic Limb, Electromyography (EMG), Deep Learning, Vision Transformer, LSTM, Reinforcement Learning, Federated Learning, Adaptive Muscle Signal Learning, Edge AI, Assistive Technology

Abstract

EMG-based prosthetic control systems are known to suffer from muscle fatigue, electrode displacement, signal drift, and inter-subject variability issues, compromising their robustness and long-term stability. While recently developed deep learning techniques deliver highly competitive gesture recognition accuracy, most existing solutions lack adaptive learning and privacy-preserving model optimization mechanisms required for the real-world implementation. In this paper, we propose a new EMG-based adaptive prosthetic control system combining the Hybrid CNN, ViT, and LSTM architectures and applying federated reinforcement learning (FRL). The CNN component of the proposed architecture extracts local spatial information from multi-channel EMG signals, the Vision Transformer analyzes inter-channel dependencies using a self-attention mechanism, while the LSTM models the temporal dynamics of muscle activation. Additionally, the reinforcement learning algorithm constantly updates control policies according to user feedback, while federated learning allows users to collaborate on model optimization without exposing raw EMG data, thus protecting user privacy. In order to enhance the system robustness in real-world conditions, the Adaptive Muscle Signal Learning (AMSL) technique is used to counteract the adverse effect of signal drift, muscle fatigue, and electrode displacement. An EMG dataset specification, called EMGPro-2026, is introduced to establish the criteria of data acquisition, gesture recognition classes, preprocessing, and evaluation process. Based on computational assessment and simulation analysis, there is proof of concept for the proposed framework that enables high gesture classification accuracy, adaptation for personalization, and efficient learning with privacy preservation for real-time use in prosthetics. The proposed CNN-ViT-LSTM with federated reinforcement learning presents an intelligent solution for the future generation of myoelectric prosthetic control systems. Our future work will entail the validation of our framework using EMG signals recorded ethically. Because the current research provides a conceptual framework, there has been no experimental validation of the performance using human EMG datasets. However, based on the benchmarks developed from the existing state-of-the-art literature, it can be said that the suggested framework for Hybrid CNN–ViT–LSTM with Federated Reinforcement Learning is predicted to deliver approximately 94.5% gesture classification accuracy, which will be better than the conventional techniques such as CNN, CNN–LSTM, and SVM and will also increase adaptability, personalization, and privacy preservation.

Cite This Paper

P. S. Saritha, S. Poonguzhali, B. Mohan, "Adaptive Myoelectric Prosthetic Control Using Hybrid CNN–Vision Transformer–LSTM Networks and Federated Reinforcement Learning", International Journal of Engineering and Manufacturing (IJEM), Vol.16, No.5, pp. 272-297, 2026. DOI:10.5815/ijem.2026.05.15

Reference

[1]Farina, D., &Aszmann, O. (2014). Bionic limbs: clinical reality and academic promises. Science Translational Medicine, 6(257), 257ps12.DOI: 10.1126/scitranslmed.3010453
[2]Jiang, N., Dosen, S., Müller, K. R., & Farina, D. (2012). Myoelectric control of artificial limbs—Is there a need to change focus? [In the Spotlight] IEEE Signal Processing Magazine, 29(5), 148–152. DOI: 10.1109/MSP.2012.2203480
[3]Krasoulis, A., Vijayakumar, S. and Nazarpour, K., 2015, April. Evaluation of regression methods for the continuous decoding of finger movement from surface EMG and accelerometry. In 2015 7th international IEEE/EMBS conference on neural engineering (NER) (pp. 631-634). IEEE. DOI: 10.1109/NER.2015.7146702
[4]Atzori, M. and Müller, H., 2015. Control capabilities of myoelectric robotic prostheses by hand amputees: a scientific research and market overview. Frontiers in systems neuroscience, 9, p.162. DOI: 10.3389/fnsys.2015.00162
[5]Ameri, A., Kamavuako, E. N., Scheme, E. J., Englehart, K. B., & Parker, P. A. (2014). Support vector regression for improved real-time, simultaneous myoelectric control.IEEE Transactions on Neural Systems and Rehabilitation Engineering, 22(6), 1198–1209. DOI: 10.1109/TNSRE.2014.2323576
[6]Phinyomark, A., Khushaba, R.N. and Scheme, E., 2018. Feature extraction and selection for myoelectric control based on wearable EMG sensors. Sensors, 18(5), p.1615. DOI: 10.3390/s18051615
[7]Ozdemir, M.A., Kisa, D.H., Guren, O., Onan, A. and Akan, A., 2020, November. EMG based hand gesture recognition using deep learning. In 2020 Medical Technologies Congress (TIPTEKNO) (pp. 1-4). IEEE. DOI: 10.1109/TIPTEKNO50054.2020.9299264
[8]Côté-Allard, U., Fall, C.L., Drouin, A., Campeau-Lecours, A., Gosselin, C., Glette, K., Laviolette, F. and Gosselin, B., 2019. Deep learning for electromyographic hand gesture signal classification using transfer learning. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 27(4), pp.760–771. DOI: 10.1109/TNSRE.2019.2896269
[9]Xiong, D., Zhang, D., Zhao, X. and Zhao, Y., 2021. Deep learning for EMG-based human-machine interaction: A review. IEEE/CAA Journal of AutomaticaSinica, 8(3), pp.512-533. DOI:10.1109/JAS.2021.1003865
[10] Simao, M., Mendes, N., Gibaru, O. and Neto, P., 2019. A review on electromyography decoding and pattern recognition for human-machine interaction. Ieee Access, 7, pp.39564-39582. DOI: 10.1109/ACCESS.2019.2906584
[11]Schmidhuber, J., 2015. Deep learning in neural networks: An overview. Neural networks, 61, pp.85-117. DOI: 10.1016/j.neunet.2014.09.003
[12]Dosovitskiy, A., et al. (2020). An image is worth 16×16 words: Transformers for image recognition at scale.arXiv:2010.11929. DOI: 10.48550/arXiv.2010.11929
[13] Z. Zhang et al., “Transformer-based biomedical signal processing: A review,” IEEE Rev. Biomed. Eng., vol. 16, pp. 210–223, 2023, DOI: 10.1109/RBME.2023.3237662
[14]Hochreiter, S. and Schmidhuber, J., 1997. Long short-term memory. Neural computation, 9(8), pp.1735-1780. DOI: 10.1162/neco.1997.9.8.1735
[15]He, K., Zhang, X., Ren, S. and Sun, J., 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778). DOI: 10.1109/CVPR.2016.90
[16]Yang, Q., Liu, Y., Chen, T. and Tong, Y., 2019. Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 10(2), pp.1-19. DOI: 10.1145/3298981
[17]Kairouz, P. and McMahan, H.B., 2021. Advances and open problems in federated learning. Foundations and trends in machine learning, 14(1-2), pp.1-210. DOI: 10.1561/2200000083
[18]Rieke, N., Hancox, J., Li, W., Milletari, F., Roth, H.R., Albarqouni, S., Bakas, S., Galtier, M.N., Landman, B.A., Maier-Hein, K. and Ourselin, S., 2020. The future of digital health with federated learning. NPJ digital medicine, 3(1), p.119. DOI: 10.1038/s41746-020-00323-1
[19]Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D. and Batra, D., 2017. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision (pp. 618-626). DOI: 10.1109/ICCV.2017.74
[20] Hinton, G., Vinyals, O. and Dean, J., 2015. Distilling the knowledge in a neural network. arXiv:1503.02531. DOI: 10.48550/arXiv.1503.02531
[21]Han, S., Mao, H. and Dally, W.J., 2015. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. arXiv:1510.00149. DOI: 10.48550/arXiv.1510.00149
[22]Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., &Bengio, Y. (2016). Quantized neuralnetworks.arXiv:1609.07061.DOI: 10.48550/arXiv.1609.07061
[23]LeCun, Y., Bengio, Y. and Hinton, G., 2015. Deep learning. nature, 521(7553), pp.436-444. DOI: 10.1038/nature14539
[24]Khan, S., Naseer, M., Hayat, M., Zamir, S.W., Khan, F.S. and Shah, M., 2022. Transformers in vision: A survey. ACM computing surveys (CSUR), 54(10s), pp.1-41. DOI: 10.1145/3505244.
[25]Geng, Y., Zhou, P. and Li, G., 2012. Toward attenuating the impact of arm positions on electromyography pattern-recognition based motion classification in transradial amputees. Journal of NeuroEngineering and Rehabilitation, 13(1), p.74. DOI: 10.1186/1743-0003-9-74
[26]Atzori, M., Cognolato, M., & Müller, H. (2016). Deep Learning with Convolutional Neural Networks Applied to Electromyography Data: A Resource for the Classification of Movements for Prosthetic Hands.Frontiers in Neurorobotics, 10, 9.DOI: 10.3389/fnbot.2016.00009
[27]ElMohandes, H., ElGayar, N., Taylor, N., Turcanu, A., Amelin, D. and Ruff, R., 2024.Enhancing myoelectric prosthetic control: Deep learning strategies for continuous arm kinematics estimation and cross-subject model transferability from EMG data.In 2024 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 1–6. IEEE.DOI: 10.1109/EMBC53108.2024.10782260
[28]Zbinden, J., Molin, J. and Ortiz-Catalan, M., 2024.Deep learning for enhanced prosthetic control: Real-time motor intent decoding for simultaneous control of artificial limbs.IEEE Transactions on Neural Systems and Rehabilitation Engineering, 32, pp.1177–1186.DOI: 10.1109/TNSRE.2024.3371896
[29]Jiang, X., Ma, C. and Nazarpour, K., 2024.Posture-invariant myoelectric control with self-calibrating random forests.
Frontiers in Neurorobotics, 18, 1462023.DOI: 10.3389/fnbot.2024.1462023
[30]Ye, Y., He, Y., Pan, T., Dong, Q., Yuan, J. and Zhou, W., 2023.Cross-subject EMG hand gesture recognition based on dynamic domain generalization.In 2023 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 1–4. IEEE.DOI: 10.1109/EMBC40787.2023.10340691
[31]Li, J., Jiang, X., Fan, J., Geng, Y., Jia, F. and Dai, C., 2025.Deep end-to-end transfer learning for robust inter-subject and inter-day hand gesture recognition using surface EMG.Biomedical Signal Processing and Control, 100, p.106892. DOI: 10.1016/j.bspc.2024.106892.
[32]S. Zabihi, E. Rahimian, A. Asif and A. Mohammadi, “TraHGR: Transformer for hand gesture recognition via electromyography,”IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 4211–4224, 2023,doi: 10.1109/TNSRE.2023.3324252.
[33]M. A. Garenfeld, M. Strbac, N. Jorgovanovic, J. L. Dideriksen and S. Dosen,“Closed-loop control of a multifunctional myoelectric prosthesis with full-state anatomically congruent electrotactilefeedback,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 2090–2100, 2023,doi: 10.1109/TNSRE.2023.3267273.
[34]M. R. K. Kadavath, M. Nasor and A. Imran,“Enhanced hand gesture recognition with surface electromyogram and machine learning,”Sensors, vol. 24, no. 16, p. 5231, 2024,doi: 10.3390/s24165231.
[35]Y. Xiang, W. Zheng, J. Tang, Y. Dong and Y. Pang,“Gesture recognition from surface electromyography signals based on the SE-DenseNetnetwork,”Biomedical Engineering / Biomedizinische Technik, vol. 70, no. 3, pp. 207–216, 2025,doi: 10.1515/bmt-2024-0282.
[36]Shehata, A. W., Williams, H. E., Hebert, J. S., &Pilarski, P. M. (2021).Machine learning for the control of prosthetic arms: Using electromyographic signals for improved performance. IEEE Signal Processing Magazine, 38(4), 46–53. DOI: 10.1109/MSP.2021.3075931.
[37]Karrenbach, M., Preechayasomboon, P., Sauer, P., Boe, D., &Rombokas, E. (2022).Deep learning and session-specific rapid recalibration for dynamic hand gesture recognition from EMG.Frontiers in Bioengineering and Biotechnology, 10, 1034672.DOI: 10.3389/fbioe.2022.1034672.
[38]Freitag, K., Karayiannidis, Y., Zbinden, J., &Laezza, R. (2026). Fine-Tuning Myoelectric Control Through Reinforcement Learning in a Game Environment. IEEE Transactions on Biomedical Engineering, 73(1), 168–179.
DOI: 10.1109/TBME.2025.3578855
[39]Berman, J., Lee, I.-C., Yin, J., & Huang, H. (2024).An Efficient Framework for Personalizing EMG-Driven Musculoskeletal Models Based on Reinforcement Learning. IEEE Transactions on Neural Systems and Rehabilitation Engineering. DOI: 10.1109/TNSRE.2024.3483150
[40]Khushaba, R. N., Al-Timemy, A. H., Kodagoda, S., &Nazarpour, K. (2017).Combined influence of forearm orientation and muscular contraction on EMG pattern recognition.Expert Systems with Applications, 61, 154–161.DOI: 10.1016/j.eswa.2016.05.031

International Journal of Engineering and Manufacturing (IJEM)