Evaluation of Random Forest and LSTM for sEMG-Based Estimation of Two DOF Wrist Kinematics
DOI:
https://doi.org/10.62535/b03ak471Keywords:
surface electromyography, wrist rehabilitation exoskeleton, Random Forest, Long short-term memory, continuous motion estimation, two-degree-of-fredoom (2-DOF)Abstract
Robotic wrist rehabilitation after stroke requires control signals that reflect the user’s volitional intent, while clinically deployable systems are constrained by the number of surface electromyography (sEMG) channels that can be practically placed on a hemiparetic forearm. This study evaluates how far a single sEMG channel can support discrete motion-intent classification and continuous two-degree-of-freedom (2-DOF) wrist kinematic estimation for exoskeleton control. Two anonymised datasets, recorded at 200 Hz using one forearm sEMG channel and one inertial measurement unit (IMU), were processed through detrending, 50 Hz notch filtering, fourth-order zero-phase 20–90 Hz band-pass filtering, and 250 ms windowing with 50% overlap. Seven time- and frequency-domain features were extracted and provided to Random Forest (RF), long short-term memory (LSTM), and hybrid RF–LSTM models. Under a strict held-out-trial protocol, LSTM achieved higher five-class classification accuracy than RF (95.56% versus 87.28%). Regression performance was weaker: LSTM produced the best flexion–extension result (R² = 0.290, RMSE = 12.49°), while hybrid RF–LSTM performed best for radial–ulnar estimation (R² = 0.204, RMSE = 12.35°). Closed-loop Dynamixel MX-28AT tracking showed zero overshoot. Overall, single-channel sEMG supports reliable intent classification, but proportional angle estimation remains limited, indicating that further sensing or regression-specific feature optimization is required for precise control.
References
Khan MA, Fares H, Ghayvat H, Brunner IC, Puthusserypady S, Razavi B, Lansberg MG, Poon A, Meador KJ. A systematic review on functional electrical stimulation based rehabilitation systems for upper limb post-stroke recovery. Front Neurol. 2023;14:1272992.
Wang X, Fang Y, Zhang Z, Zhao X, Xiong D, Li J. Review of soft robotic gloves and functional electrical stimulation affecting hand function rehabilitation for stroke patients. Biomimetics. 2026;11(2):104.
Fugl-Meyer AR, Jääskö L, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient. 1. A method for evaluation of physical performance. Scand J Rehabil Med. 1975;7(1):13–31.
Brunnstrom S. Motor testing procedures in hemiplegia based on sequential recovery stages. Phys Ther. 1966;46(4):357–375.
Campanini I, Disselhorst-Klug C, Rymer WZ, Merletti R. Surface EMG in clinical assessment and neurorehabilitation: barriers limiting its use. Front Neurol. 2020;11:934.
Hudgins B, Parker P, Scott RN. A new strategy for multifunction myoelectric control. IEEE Trans Biomed Eng. 1993;40(1):82–94.
Phinyomark A, Quaine F, Charbonnier S, Serviere C, Tarpin-Bernard F, Laurillau Y. EMG feature evaluation for improving myoelectric pattern-recognition robustness. Expert Syst Appl. 2013;40(12):4832–4840.
Parajuli N, Sreenivasan N, Bifulco P, Cesarelli M, Savino S, Niola V, Esposito D, Hamilton TJ, Naik GR, Gunawardana U, Gargiulo GD. Real-time EMG-based pattern-recognition control for hand prostheses: a review of existing methods, challenges, and future implementation. Sensors. 2019;19(20):4596.
Hahne JM, Biessmann F, Jiang N, Rehbaum H, Farina D, Meinecke F, Müller KR, Parra L. Linear and nonlinear regression techniques for simultaneous and proportional myoelectric control. IEEE Trans Neural Syst Rehabil Eng. 2014;22(2):269–279.
Ameri A, Akhaee MA, Scheme E, Englehart K. Regression convolutional neural network for improved simultaneous EMG control. J Neural Eng. 2019;16(3):036015.
Breiman L. Random forests. Mach Learn. 2001;45(1):5–32.
Xiao F, Wang Y, Gao Y, Zhu Y, Zhao J. Continuous estimation of joint angle from electromyography using multiple time-delayed features and random forests. Biomed Signal Process Control. 2018;39:303–311.
Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997;9(8):1735–1780.
Ma C, Lin C, Samuel OW, Xu L, Li G. Continuous estimation of upper-limb joint angle from sEMG signals based on an SCA-LSTM deep-learning approach. Biomed Signal Process Control. 2020;61:102024.
Zhang Y, Zhang X, Lu Z, Jiang Z, Zhang T. A novel wrist-joint torque prediction method based on EMG and LSTM. In: Proc. Int. Conf. Cyber Technology in Automation, Control, and Intelligent Systems. 2020. p. 242–245.
Bao T, Zaidi SAR, Xie S, Yang P, Zhang ZQ. A CNN-LSTM hybrid model for wrist kinematics estimation using surface electromyography. IEEE Trans Instrum Meas. 2021;70:2503809.
Du J, Liu Z, Dong W, Zhang W, Miao Z. A novel TCN-LSTM hybrid model for sEMG-based continuous estimation of wrist joint angles. Sensors. 2024;24(17):5631.
Wang H, Tao Q, Zhang X. Ensemble-learning method for the continuous decoding of hand joint angles. Sensors. 2024;24(2):660.
Wei Z, Zhang ZQ, Xie SQ. Continuous motion intention prediction using sEMG for upper-limb rehabilitation: a systematic review of model-based and model-free approaches. IEEE Trans Neural Syst Rehabil Eng. 2024;32:1487–1504.
Yang W, Yang D, Liu Y, Liu H. Decoding simultaneous multi-DOF wrist movements from raw EMG signals using a convolutional neural network. IEEE Trans Hum-Mach Syst. 2019;49(5):411–420.
Liu J, Ren Y, Xu D, Kang SH, Zhang LQ. EMG-based real-time linear-nonlinear cascade regression decoding of shoulder, elbow, and wrist movements in able-bodied persons and stroke survivors. IEEE Trans Biomed Eng. 2020;67(5):1272–1281.
Kumar S, Wöhrle H, Trampler M, Simnofske M, Peters H, Mallwitz M, Kirchner EA, Kirchner F. Modular design and decentralized control of the Recupera exoskeleton for stroke rehabilitation. Appl Sci. 2019;9(4):626.
ROBOTIS Co., Ltd. MX-28T/R/AT/AR(2.0). ROBOTIS e-Manual. 2026 [accessed 7 Aug 2026].
Zhao Y, Qian K, Sheng B, Zhang ZQ, Li Z, Li GQ, Dehghani-Sanij AA, Xie SQ. Adaptive cooperative control strategy for a wrist exoskeleton using model-based joint impedance estimation. IEEE/ASME Trans Mechatronics. 2023;28(2):748–757.
Zheng Z, et al. Adaptive compliance control for a wearable lower-limb rehabilitation robot based on online estimation of the user's joint stiffness. IEEE/ASME Trans Mechatronics. 2025;30(6):5590–5602.
Thorat M, Mehta R, Nayyar M, Oza R, Mane R. An EMG-based expert system for detection of wrist flexion and extension. In: Proc. 3rd Int. Conf. Futuristic Technologies (INCOFT). 2025. p. 278–285.
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–2830.
Kingma DP, Ba J. Adam: a method for stochastic optimization. In: Proc. Int. Conf. Learning Representations (ICLR). 2015.
Selvan MP, Raj R, Sai RG, Jancy S, Mary VA. Prosthetic hand using EMG. J Phys Conf Ser. 2021;1770(1):012018.
van der Maaten L, Hinton G. Visualizing data using t-SNE. J Mach Learn Res. 2008;9:2579–2605.
