Shi Jun, Zheng Yong-Ping, Huang Qing-Hua, Chen Xin
Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Hong Kong.
IEEE Trans Biomed Eng. 2008 Mar;55(3):1191-8. doi: 10.1109/TBME.2007.909538.
The aim of this study is to demonstrate the feasibility of using the continuous signals about the thickness and pennation angle changes of muscles detected in real-time from ultrasound images, named as sonomyography (SMG), to characterize muscles under isometric contraction, along with synchronized surface electromyography (EMG) and generated torque signals. The right biceps brachii muscles of seven normal young adult subjects were tested. We observed that exponential functions could well represent the relationships between the normalized EMG root-mean-square (RMS) and the torque, the RMS and the muscle deformation SMG, and the RMS and the pennation angle SMG for the data of the contraction phase, with exponent coefficients of 0.0341 +/- 0.0148 (Mean SD), 0.0619 +/- 0.0273, and 0.0266 +/- 0.0076, respectively. In addition, the preliminary results also demonstrated linear relationships between the normalized torque and the muscle deformation as well as the pennation angle with the ratios of 9 .79 +/- 3.01 and 2.02 +/- 0.53, respectively. The overall mean R2 for the regressions was approximately 0.9 and the overall mean relative root mean square error (RRMSE) smaller than 15%. The potential values of SMG together with EMG to provide a more comprehensive assessment for the muscle functions should be further investigated with more subjects and more muscle groups.
本研究的目的是证明利用超声图像实时检测到的肌肉厚度和羽状角变化的连续信号(即肌声图,SMG),结合同步表面肌电图(EMG)和产生的扭矩信号,对等长收缩下的肌肉进行特征描述的可行性。对7名正常年轻成年受试者的右侧肱二头肌进行了测试。我们观察到,对于收缩期数据,指数函数能够很好地表示归一化肌电图均方根(RMS)与扭矩、RMS与肌肉变形SMG以及RMS与羽状角SMG之间的关系,指数系数分别为0.0341±0.0148(均值±标准差)、0.0619±0.0273和0.0266±0.0076。此外,初步结果还表明,归一化扭矩与肌肉变形以及羽状角之间存在线性关系,比值分别为9.79±3.01和2.02±0.53。回归的总体平均R2约为0.9,总体平均相对均方根误差(RRMSE)小于15%。应通过更多受试者和更多肌肉群进一步研究SMG与EMG结合为肌肉功能提供更全面评估的潜在价值。