Research on Personalized FES Regulation Based on sEMG GDSNTF
Du Yihao*, Wang Xiaoran, Sun Mengyu, Li Jingjin, Wu Xiaoguang
(Key Laboratory of Intelligent Control and Neural Information Processing, Ministry of Education, Key Laboratory of Intelligent Rehabilitation and Neuromodulation of Hebei Province, Key Laboratory of Test & Measurement Technology and Instrumentation of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, Hebei, China)
Abstract：Functional electrical stimulation (FES) is widely used in fields of motor recovery and neurological rehabilitation. To improve rehabilitation efficiency and achieve personalized rehabilitation, and to address the issues of real-time performance and stability in FES control, this study proposed a model based on gradient descent optimized L1/2-norm sparsenon-negative tensor factorization (GDSNTF). This model simultaneously extracted muscle synergy information in three dimensions: temporal, spatial, and frequency domains, enabling a comprehensive and in-depth exploration of muscle synergy characteristics during human movement. The GDSNTF model was applied to the regulation of FES stimulation parameters. By introducing a weight optimization algorithm for primary motor patterns, the real-time performance and stability of FES control were improved, and weighted muscle synergies were used to enhance the personalization of FES stimulation. Ten healthy adult males were recruited as subjects, and a hand movement mirror control experiment was designed. Four movements were selected: wrist flexion, wrist extension, fist clenching, and palm opening. Surface electromyography (sEMG) signals were collected from the unaffected hand during movement, and muscle synergy features (temporal, spatial, and frequency domains) were extracted using GDSNTF to regulate FES stimulation parameters (current intensity, amplitude, and frequency) on the affected side. Comparative analyses were performed with non-negative tensor factorization (NTF) and sparse non-negative tensor factorization (SNTF). One-way analysis of variance (ANOVA) was used to verify the effectiveness of the personalized FES control method based on GDSNTF. The results showed that after 20 iterations, the relative errors of NTF and SNTF were 0.30±0.03 and 0.29±0.02, respectively, whereas that of GDSNTF was 0.20±0.02, with highly significant differences among groups (F=68.52, P<0.01), indicating higher accuracy of the GDSNTF algorithm. Meanwhile, the runtime of GDSNTF was (2.10±0.15) s, significantly shorter than (2.67±0.21) s for NTF and (2.48±0.18) s for SNTF, with highly significant inter-group differences (F=42.36, P<0.01). In conclusion, the proposed method achieved superior real-time performance while ensuring accuracy.
杜义浩, 王孝冉, 孙梦雨, 李菁金, 吴晓光. 基于sEMG-GDSNTF的个性化FES调控研究[J]. 中国生物医学工程学报, 2026, 45(3): 312-320.
Du Yihao, Wang Xiaoran, Sun Mengyu, Li Jingjin, Wu Xiaoguang. Research on Personalized FES Regulation Based on sEMG GDSNTF. Chinese Journal of Biomedical Engineering, 2026, 45(3): 312-320.
[1] 《中国脑卒中防治报告2021》概要[J].中国脑血管病杂志,2023,20(11):783-793. [2] Iwamoto Y,Imura T,Suzukawa T,et al. Combination of exoskeletal upper limb robot and occupational therapy im-prove activities of daily living function in acute stroke pa-tients[J]. J Stroke Cerebrovasc Dis,2019,28(7):2018-2025. [3] 姜畅,吴建贤. 虚拟现实技术在脑卒中后上肢功能康复中的研究进展[J]. 中华临床医师杂志(电子版),2017,11(9):1627-1630. [4] Sue-Min L, Stephanie SWPD. Persisting consequences of stroke measured by the Stroke Impact Scale.[J]. Stroke, 2002,33(7):1840-1844. [5] 郭晓辉, 王晶, 徐光华. 手部功能康复机器人研究最新进展 [J]. 中国康复医学杂志, 2017, 32(2): 235-240. [6] 刘林林,魏冬梅,苏庆文,等. 低频重复经颅磁刺激联合Bobath疗法对脑卒中偏瘫患者神经功能、运动功能及平衡能力的影响[J].现代生物医学进展,2023(6):126-130. [7] Peckham HP, Knutson SJ. Functional electrical stimulation for neuromuscular applications [J]. Annual Review of Biomedical Engineering,2005(7):327-360. [8] Gater DR, Dolbow D, Tsui B, et al. Functional electrical stimulation therapies after spinalcord injury[J]. NeuroRehabilitation, 2011, 28(3): 231-248. [9] Sujith OK. Functional electrical stimulation in neurological disorders[J]. European Journal of Neurology, 2010,15(5): 437-444. [10] 梁成盼,丁文娟,黄桂兰,等.多通道功能性电刺激对脑卒中上肢运动功能的疗效及表面肌电特征研究[J].中国康复医学杂志,2021,36(9):1127-1130. [11] Juliana M, Cláudia S. Usability of functional electrical stimulation in upper limb rehabilitation in post-stroke patients: a narrative review[J].Sensors,2022,22(4):1409-1409. [12] Karamanavi E. The FES gene, located at the chromosome 15Q21.6 coronary-artery-disease locus, modulates atherosclerotic plaque vulnerability[J]. Atherosclerosis, 2020,315: e19-e19. [13] Lee DD, Seung HS. Learning the parts of objects by non-negative matrix factorization[J]. Nature, 1999, 401(6755): 788-791. [14] Shashua A, Hazan T. Non-negative tensor factorization with applications to statistics and computer vision[C]//Proceedings of the 22nd International Conference on Machine learning. New York: Association for Computing Machinery, 2005: 792-799. [15] 宋珊. 基于矩阵分解与张量分解的统计模型及应用[D]. 苏州:苏州科技大学, 2021. [16] 陈妙慧. 基于非负正交约束的大规模Tucker分解算法[D]. 广州;华南理工大学, 2021. [17] Yin W, Qu Y, Ma Z, et al. HyperNTF: a hypergraph regularized nonnegative tensor factorization for dimensionality reduction[J]. Neurocomputing, 2022, 512: 190-202. [18] Lin C. Projected gradient methods for nonnegative matrix factorization[J]. Neural Computation, 2007, 19(10): 2756-2779. [19] Mehrabi N, Schwartz MH, Steele KM. Can altered muscle synergies control unimpaired gait?[J]. Journal of Biomechanics, 2019, 90: 84-91. [20] 杨毅,彭玉鑫,郝增明,等. 复杂人体运动冗余控制的肌肉协同理论研究进展与展望[J]. 体育科学, 2020,40: 63-72. [21] Tresch MC, Cheung VC, D'Avella A. Matrix factorization algorithms for the identification of muscle synergies: evaluation on simulated and experimental data sets[J]. Journal of Neurophysiology, 2006, 95(4): 2199-2212. [22] Ebied A, Kinney-Lang E, Spyrou L, et al. Evaluation of matrix factorisation approaches for muscle synergy extraction[J]. Medical Engineering & Physics, 2018, 57: 51-60.
