In order to investigate the dynamic coupling mechanism between central regulation and peripheral responses, a three-dimensional electromyography (EMG) fatigue evaluation model for upper limb multi-joint movements was developed using a brain-computer interface (BCI) training approach. Surface EMG signals from four channels, including the brachioradialis muscle, were recorded from 30 participants during four types of upper limb movements (fist clenching, wrist flexion, wrist elevation, and elbow flexion). A comprehensive evaluation system integrating subjective perception, peripheral physiology, and central performance was constructed by combining the fatigue scale-14 (FS-14) subjective fatigue scale with common spatial pattern classification accuracy.The results are as follows. Significant progressive decreases are observed in time-domain features across three stages (p<0.05), although compensatory enhancement is found in specific channels. Dynamic changes in high-frequency energy distribution are demonstrated by time-frequency analysis, which confirms the preferential recruitment mechanism of type IIb fast-twitch muscle fibers. The optimal contributions are determined through entropy-based weighting for subjective fatigue, system performance, and peripheral physiology. The composite fatigue index exhibits significant progressive increases across stages (p<0.01). It is indicated that the model effectively integrates multi-dimensional fatigue signals, providing new insights for dynamic monitoring and rehabilitative interventions in neuromuscular fatigue.Experimental data indicate that greater classification accuracy fluctuations (maximum variation 0.13) are demonstrated in movement execution tasks compared to motor imagery tasks. It is suggested that BCI system designs require task-specific fatigue intervention strategies.
LI Zhao-lin, CHAI Xiao-ke, QI Kai-cheng, SHAN Xin-ying, GUO Shi-jie.
Three-dimensional Dynamic Assessment Model for Upper Limb Fatigue Based on Brain-computer Interface and Surface Electromyography[J].
Science Technology and Engineering, 2026, 26(7): 2667-2677 DOI:10.12404/j.issn.1671-1815.2502832
运动疲劳作为神经肌肉系统功能衰退的综合性表征,既是运动医学研究的核心命题,也是脑机接口(brain-computer interface,BCI)系统效能衰减的关键诱因[3]。传统疲劳评估多依赖主观量表(疲劳量表-14)或力学性能测试[4-5],这些方法虽能反映整体疲劳感知,却难以揭示中枢调控与外周响应的动态耦合机制。近年来,人工智能技术的突破推动了多维信号分析发展:Lin等[6]基于BERT(bidirectional encoder representations from transformers)模型实现了跨被试肌电解码,有效提升了个体差异适应性;Yuan等[7]结合表面肌电(surface electromyography,sEMG)与惯性传感器优化了手势识别精度。陈志航等[8]进一步提出跨模态注意力机制,显著提升了脑肌电信号融合解码效果。然而,这些方法仍局限于静态场景,未建立疲劳状态下的动态耦合评估框架。刘光达等[9]近期提出的果蝇-遗传优化神经网络(fruit fly-genrtic optimization algorithm-neural network,FGNN),通过融合智能算法显著提升了肌肉疲劳的分类精度,为动态评估提供了新工具。辛运胜等[10]基于表面肌电实验对挖掘机驾驶员上肢肌肉疲劳的分析,进一步揭示了在实际动态操作环境中肌肉疲劳的特征,为理解复杂任务条件下的疲劳机制提供了重要参考。随着sEMG信号采集技术的突破性进展,研究者得以通过非侵入方式获取运动单元动作电位(motor unit action potential,MUAP)的时空编码特征[11],为解析“脑-脊髓-肌肉”通路的疲劳传导机制提供了全新视角。
近年来,时域分析方法因其计算高效性,被广泛用于局部肌肉负荷评估[12]。Favretto等[13]证实,糖尿病患者的胫骨前肌在等距收缩中均方根(root mean square,RMS)值随时间显著增加,且II型肌纤维比例较高。这表明II型肌纤维募集策略的改变(如早期疲劳)可能导致RMS值的非线性增长。然而,sEMG信号的个体差异性[14-15]以及运动想象(motor imagery,MI)任务中中枢-外周信号传导的异步性[16-18],严重制约了传统时域特征在复杂任务中的普适性。与此同时,频域分析通过中值频率(median frequency,MF)偏移现象揭示了肌肉疲劳的代谢性本质——乳酸堆积导致的肌膜传导速度下降可使MF值每小时衰减[19]。但这一理论难以解释前期研究发现的高频补偿现象,提示现有模型可能低估了中枢神经系统的动态代偿能力。Zhao等[20]通过速度损失量化证实Ⅱ型肌纤维代谢衰退早于慢肌纤维,张雨薇等[21]结合fNIRS技术验证了卒中患者疲劳后期的中枢代谢代偿现象,而Shen等[22]指出BCI系统在长期康复训练中因疲劳累积导致依从性下降,二者共同凸显了中枢代偿机制解析的紧迫性。李伟明等[23]虽开发了肌电反馈闭环康复系统,但未解决BCI系统疲劳衰减问题,为神经肌肉疲劳的动态监测与康复干预提供了新思路。
TangWeidi. Research on key technologies of upper limb movement analysis based on surface electromyography[D]. Beijing: University of Science and Technology of China, 2021.
[3]
张政委. 基于表面肌电的短视频用户上肢肌肉疲劳特性研究[D]. 天津: 天津科技大学, 2020.
[4]
ZhangZhengwei. Research on upper limb muscle fatigue characteristics of short video users based on surface electromyography[D]. Tianjin: Tianjin University of Science and Technology, 2020.
ZhangLi, ZhangXiaolin, ZhaiXiaowei, et al. Effect of transcranial ultrasound neuromuscular electrical stimulation combined with brain-computer interface training on upper limb function and quality of life in stroke patients with hemiplegia[J]. Journal of Modern Medicine and Health Research, 2025, 9(5): 48-50.
XuHua, XiongBo, ChenJingjing, et al. Correlation study between FS-14 and FAI assessments in helicopter pilots of the army aviation force[J]. Journal of Clinical Psychosomatic Diseases, 2023, 29(6): 77-79, 90.
[9]
PethickJ, TallentJ. The neuromuscular fatigue-induced loss of muscle force control[J]. Sports, 2022, 10(11): 184.
[10]
LinC, ChenX, GuoW, et al. A BERT based method for continuous estimation of cross-subject hand kinematics from surface electromyographic signals[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2023, 31: 87-96.
[11]
YuanY, LiuJ, DaiC, et al. Exploring pattern-specific components associated with hand gestures through different sEMG measures[J]. Journal of NeuroEngineering and Rehabilitation, 2024, 21(1): 1-13.
ChenZhihang, WangLiqun, ZhangFan, et al. A brain-muscle signal decoding model based on cross-modal attention mechanism[J]. Chinese Journal of Biomedical Engineering, 2024, 43(2): 145-152.
LiuGuangda, XuLanyu, XiaoRuolan, et al. Muscle fatigue classification method of sEMG signal based on FGNN algorithm[J]. Science Technology and Engineering, 2022, 22(19): 8370-8377.
XinYunsheng, BaiWei, JinHongyi, et al. Fatigue analysis of upper limb muscle of excavator driver based on myoelectric experiment[J]. Science Technology and Engineering, 2024, 24(26): 11393-11400.
[18]
王鹏. 基于脑机接口的精神疲劳评估方法研究[D]. 重庆: 重庆大学, 2011.
[19]
WangPeng. Research on mental fatigue assessment method based on brain-computer interface[D]. Chongqing: Chongqing University, 2011.
MaJing, LiuGongju, WangJian. Research and prospect on array surface electromyography characteristics of muscle fatigue[J]. Chinese Journal of Physical Medicine and Rehabilitation, 2018, 40(4): 318-320.
[22]
FavrettoM A, CossulS, AndreisF R, et al. Alterations of tibialis anterior muscle activation pattern in subjects with type 2 diabetes and diabetic peripheral neuropathy[J]. Biomedical Physics & Engineering Express, 2022, 8(2): ac455b.
WangKunpeng, PangJie, ShiLei, et al. Methods for feature extraction of surface electromyography signals of human leg[J]. Journal of Chongqing University, 2017, 40(11): 83-90.
[25]
YuanY, LiuJ, DaiC, et al. Exploring pattern-specific components associated with hand gestures through different sEMG measures[J]. Journal of NeuroEngineering and Rehabilitation, 2024, 21(1): 1-13.
LiuTiejun, XuPeng, YuQian, et al. Research progress on the brain mechanism of motor imagery and its application in motor function rehabilitation[J]. Progress in Biochemistry and Biophysics, 2011, 38(4): 299-304.
[28]
ShenS, ChuT, WangJ, et al. Progress in the application of motor imagery therapy in upper limb motor function rehabilitation of stroke patients with hemiplegia[J]. Frontiers in Neurology, 2025, 16: 1454499.
[29]
LiuG, ZhangR, TianL, et al. Fine-grained spatial-frequency-time framework for motor imagery brain-computer interface[J]. IEEE Journal of Biomedical and Health Informatics, 2025, 33(2): 210-225.
ZhangSu, GuoFeng, WangNianhui, et al. Characteristic analysis of surface electromyography signals during isokinetic flexion-extension exercise inducing lower limb knee muscle fatigue[J]. Journal of Jilin University (Medical Edition), 2014, 40(6): 1241-1246.
[32]
ZhaoH, KurokawaT, TajimaM, et al. Can perceived exertion and velocity loss serve as indirect indicators of muscle fatigue during explosive back squat exercise?[J]. Journal of Functional Morphology and Kinesiology, 2024, 9(4): 238.
ZhangYuwei, LiuZhenyu, LiMing, et al. Central-peripheral coupling mechanism of upper limb fatigue in stroke patients[J]. Aerospace Medicine and Medical Engineering, 2023, 36(5): 321-328.
[35]
ShenS, ChuT, WangJ, et al. Progress in the application of motor imagery therapy in upper limb motor function rehabilitation of stroke patients with hemiplegia[J]. Frontiers in Neurology, 2025, 16: 1454499.
LiWeiming, ChenXiaoyang, ZhouTao, et al. Effects of an electromyographic feedback closed-loop rehabilitation system on upper limb functional recovery in stroke patients[J]. Chinese Journal of Physical Medicine and Rehabilitation, 2025, 47(3): 189-195.
[38]
李玉章. 表面肌电在体育中的应用[M]. 上海: 复旦大学出版社, 2015: 275.
[39]
LiYuzhang. Application of surface electromyography in sports[M]. Shanghai: Fudan University Press, 2015: 275.
[40]
FarinaD, MerlettiR, EnokaR M. The extraction of neural strategies from the surface EMG: 2004—2024[J]. Journal of Applied Physiology, 2025, 138(1): 121-135.
[41]
岑尧辉. 在线异步MI-BCI系统实验范式设计[D]. 哈尔滨: 哈尔滨工业大学, 2019.
[42]
CenYaohui. Experimental paradigm design of online asynchronous MI-BCI system[D]. Harbin: Harbin Institute of Technology, 2019.
LeiJuntian, BaiZhenmin, ZhangHanyue, et al. Progress in the application of nonlinear dynamics analysis methods in electromyography signals[J]. Chinese Journal of Sports Medicine, 2024, 43(7): 554-562.
WangKun, WangChengjun, LuoErping, et al. Research on the time and frequency domain variation rules of surface electromyography signals in muscle fatigue state based on amplitude-frequency joint analysis method[J]. Chinese Journal of Medical Physics, 2010, 27(4): 2030-2033, 2038.
[49]
Díaz-LaraJ, ReismanE, BotellaJ, et al. Delaying post-exercise carbohydrate intake impairs next-day exercise capacity but not muscle glycogen or molecular responses[J]. Acta Physiologica, 2024, 240(10): e14215.
ChenXing, YinYuehong. Dynamic system-markov model of the Active Postsynaptic response of muscle spindle afferent neurons[J]. Chinese Science Bulletin, 2013, 58(9): 793-802.
[52]
BlumK P, LamotteD, IncampsB, et al. Force encoding in muscle spindles during stretch of passive muscle[J]. PLoS computational Biology, 2017, 13(9): e1005767.
ZouYing, LiJing, FanYongmei, et al. Near-infrared brain functional imaging during motor imagery and motor execution in healthy adults[J]. Journal of Central South University (Medical Edition), 2022, 47(7): 920-927.
[55]
KobeltM, WirthB, Schuster-AmftC. Muscle activation during grasping with and without motor imagery in healthy volunteers and patients after stroke or with Parkinson's disease[J]. Frontiers in Psychology, 2018, 9: 597.
PDF
(12319KB)
