1 School of Mechanical and Electrical Engineering, Beijing Information Science and Technology University, Beijing 100192, China
2 Institute of Mechatronic Engineering, Tsinghua University, Beijing 100084, China
3 Key Laboratory of Modern Measurement and Control Technology, Beijing Information Science and Technology University, Beijing 100192, China
Show less
文章历史+
Received
Published
Revised
2025-09-14
2026-05-28
2025-11-27
Online Date
2026-08-07
PDF (8818K)
摘要
为解决单一肌肉群表面肌电(surface electromyography,sEMG)信号在康复训练评估中参考性不足的问题,通过对多肌群sEMG特征融合与筛选研究,为慢性疼痛患者与运动功能障碍群体提供高精度的量化评估方法。同步采集20名健康受试者肱二头肌、肱三头肌及肱桡肌的sEMG信号与主观疲劳评分(ratings of perceived exertion, RPE),通过小波去噪与特征提取获得时域[均方根值(root mean square,RMS)]与频域[中值频率(median frequency,MF)]特征,并分析其与RPE之间的相关性。基于Pearson相关系数筛选显著特征,构建线性回归(linear regression,LR)、支持向量机回归(support vector machine regression,SVMR)、套索回归(Lasso)及逐步回归(stepwise regression,SWR)模型对融合特征进行建模比较。结果表明:SVMR模型融合三肌肉群RMS与MF特征时性能最优,肱桡肌RMS与RPE的相关性最高(r = 0.841),均方误差(mean square error,MSE)为0.077,较单一模型降低78%,判定系数R2达0.950,评估准确率提升35%。提出的多肌群sEMG特征处理与建模方法,能有效提高康复训练评估的精度,具有良好的临床推广与应用潜力。
Abstract
In order to address the limitation that surface electromyography (sEMG) signals from a single muscle group provide insufficient reference for rehabilitation training assessment, the multi-muscle sEMG feature fusion and selection method was proposed to provide a high-precision quantitative evaluation for patients with chronic pain and motor dysfunction. sEMG signals of the biceps brachii, triceps brachii, and brachioradialis were synchronously collected from 20 healthy subjects along with their ratings of perceived exertion (RPE). Wavelet denoising and feature extraction were performed to obtain time-domain (root mean square, RMS) and frequency-domain (median frequency, MF) features, and their correlations with RPE were further analyzed. Significant features were selected based on Pearson correlation coefficients, and multiple regression models—including linear regression (LR), support vector machine regression (SVMR), Lasso regression, and stepwise regression (SWR)—were constructed and compared. Results show that the SVMR model achieved the best performance when fusing RMS and MF features of the three muscle groups, with the RMS of the brachioradialis exhibiting the highest correlation with RPE (r = 0.841). The mean square error (MSE) is 0.077, representing a 78% reduction compared with single-muscle models, while the coefficient of determination (R2) reached 0.950, corresponding to a 35% improvement in evaluation accuracy. The proposed multi-muscle sEMG feature processing and modeling method can effectively enhance the accuracy of rehabilitation training assessment and demonstrates promising potential for clinical application and translation.
YANG Xiao, SUN Jiang-hong, GAO Ke-ke, HU Fu-qing, MA Chao, ZHOU Fu-qiang.
Fusion and Selection of Multi-muscle sEMG Features for Rehabilitation Training Assessment[J].
Science Technology and Engineering, 2026, 26(15): 6317-6325 DOI:10.12404/j.issn.1671-1815.2506614
HongXiaohui, HuangLi, LiPeiling, et al. Application of individualized extended rehabilitation service in stroke patients with limb dysfunction[J]. Modern Medicine & Health, 2025, 41(2): 417-420.
[3]
XiongL H, LiX Z, LuX, et al. Impact of early refined nursing program on prognosis of middle-aged and elderly patients with cognitive dysfunction combined with cerebral infarction[J]. World Journal of Psychiatry, 2024, 14(7): 1034-1042.
TangRuirui, HuLongjun, XuJumi, et al. Investigation on status quo of the standard assistive devices agencies construction for the disabled in Shanghai[J]. Medicine and Society, 2023, 36(2): 19-23.
WeiLinmin, LuJian, YuanHong, et al. Study of the effect of the whole-process management mode of assistive devices on the activities of daily living of physically disabled people[J]. Shanghai Medical & Pharmaceutical Journal, 2024, 45(6): 42-44, 57.
ShangYuewei, TaoQing, ZhouZhiwei, et al. Development of a series rehabilitation robot for wrist joint[J]. Science Technology and Engineering, 2025, 25(27): 11676-11682.
LiaoBin, LuJiaqi, LiYu, et al. Design of an upper limb rehabilitation robot based on a cable-driven parallel mechanism[J]. Science Technology and Engineering, 2022, 22(32): 14229-14236.
[12]
ShiL, WangR, ZhaoJ, et al. Detection of rehabilitation training effect of upper limb movement disorder based on MPL-CNN[J]. Sensors, 2024, 24(4): 1105.
[13]
AleksandraR, AnnaL. The usefulness of surface electromyography in rehabilitation and physiotherapy: systematic review[J]. Pomeranian Journal of Life Sciences, 2020, 66(3): 49-56.
[14]
XueH, YanZ, FengW, et al. Musculoskeletal rehabilitation status monitoring based on sEMG[J]. Mobile Information Systems, 2021(1): 4482578.
[15]
LiuY, LiX L, ZhuA B, et al. Design and evaluation of a surface electromyography-controlled lightweight upper arm exoskeleton rehabilitation robot[J]. International Journal of Advanced Robotic Systems, 2021, 18(3): 17298814211003461.
[16]
JaehyungK, ByunggookK, YeonggwanI. Surface electromyography for evaluating patients with oromandibular dystonia[J]. CRANIO : The Journal of Craniomandibular Practice, 2021, 42(3): 1-9.
[17]
HeJ, NiuX, ZhaoP, et al. From forearm to wrist: deep learning for surface electromyography-based gesture recognition[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering : A Publication of the IEEE Engineering in Medicine and Biology Society, 2024, 32: 102-111.
[18]
YuB, ZhangX, ChengY, et al. The effects of the biceps brachii and brachioradialis on elbow flexor muscle strength and spasticity in stroke patients[J]. Neural Plasticity, 2022(1): 1295908.
[19]
GiovanniC, SilviaC. A simulation study to assess the factors of influence on mean and median frequency of sEMG signals during muscle fatigue[J]. Sensors, 2022, 22(17): 6360.
[20]
LiaoF Y, ZhangX Y, CaoC H M, et al. Effects of muscle fatigue and recovery on complexity of surface electromyography of biceps brachii[J]. Entropy, 2021, 23(8): 1036.
[21]
YunI, JeungJ, SongY H, et al. Non-invasive quantitative muscle fatigue estimation based on correlation between sEMG signal and muscle mass[J]. IEEE Access, 2020, 8: 191751-191757.
LiuGuangda, DongMengkun, XuLanyu, et al. Characteristic analysis of surface electromyography signal in arm fatigue[J]. Science Technology and Engineering, 2021, 21(25): 10690-10696.
[24]
MoniriA, TerracinaD. Real-time forecasting of sEMG features for trunk muscle fatigue using machine learning[J]. IEEE Transactions on Bio-medical Engineering, 2020, 68(2): 718-727.
[25]
RokselaA. Evaluation of neurological disorders in isokinetic dynamometry and surface electromyography activity of biceps and triceps muscles[J]. The Latest Developments and Challenges in Biomedical Engineering, 2024, 746: 151-162.
[26]
MatteoB P. Upper and lower limb performance fatigability in people with multiple sclerosis investigated through surface electromyography: a pilot study[J]. Physiological Measurement, 2020, 41(2), 457-462.
[27]
DingY, ZhouS, ZengL. Optimal parameters for rapid evaluation of muscle fatigue degree by surface electromyograph[J]. Analytical and Quantitative Cytopathology and Histopathology, 2022, 43(6): 659-668.
WangHui, HuangPingao, ZhangYuankang, et al. Muscle spasm evaluation method based on synchronous analysis of electromyography and muscle movement[J]. Journal of Instrumentation, 2019, 40 (8): 176-183.
[30]
ZhongS, JiaN, QuY, et al. Analysis and study on biomarkers of local muscle fatigue caused by repetitive lifting task[J]. BMC Musculoskeletal Disorders, 2024, 25(1): 660.
[31]
KahlL, HofmannU G. Comparison of algorithms to quantify muscle fatigue in upper limb muscles based on sEMG signals[J]. Medical Engineering & Physics, 2016, 38(11): 1260-1269.
PanLizheng, SongAiguo, XuGuozheng, et al. Prescription diagnosis of upper limb rehabilitation training robot based on SVM-GDFNN[J]. Chinese Journal of Mechanical Engineering, 2013, 49(13): 17-23.
[34]
LiuQ, LiuY, ZhangC, et al. sEMG-based dynamic muscle fatigue classification using SVM with improved whale optimization algorithm[J]. IEEE Internet of Things Journal, 2021, 8(23): 16835-16844.
[35]
ShantalM, OthmanZ, BakarA A. A novel approach for data feature weighting using correlation coefficients and min-max normalization[J]. Symmetry, 2023, 15: 2185.
[36]
NoushinF G, StefanoS, GiacomoS, et al. Assessing aberrant muscle activity patterns via the analysis of surface EMG data collected during a functional evaluation.[J]. BMC Musculoskeletal Disorders, 2019, 20(1): 13.