北京生物医学工程

基于肌动图与肌电图信号的假肢控制系统的研究

A Prosthesis Control System Based on Both of Mechanomyography and Electromyography

作者：游淼邹国栋林婉华余龙

单位：中南大学地球科学与信息物理工程学院（长沙410083）

分类号：

出版年·卷·期（页码）：2011·30·6（574-577）

摘要：

目的验证使用肌动图（mechanomyography，MMG）和肌电图（electromyography，EMG）两种信号共同作为假肢控制信号时，是否能提高假肢控制系统分类的准确度。方法本文采用信号融合方法，通过融合6通道的MMG信号与2通道的EMG信号，以及基于模式识别的线性判别分析（linear discriminant analysis，LDA）算法，研制了基于MMG和EMG信号的假肢控制系统。结果该系统能对采集到的信号进行处理并得出动作分类结果，然后控制假肢完成相应动作。对6位测试者的腕屈、腕伸、张开、握拳4类动作以及静止状态进行假肢控制的动作分类准确度实验，准确度达94.6%，比单独用MMG信号的精度88.5%或EMG信号精度90.4%效果更好。结论基于MMG与EMG信号的假肢控制系统可以更好地实现假肢控制动作的有效分类，未来可应用于上臂截肢的残疾人。

Objective To improve the accuracy of classification for a prosthesis control system by using of mechanomyography（MMG）and electromyography（EMG）as prosthesis control signals.Methods A prosthesis control system based on MMG and EMG was developed by using a signal fusion method.Six channels of MMG and two channels of EMG were fused，and combined with linear discriminant analysis（LDA）algorithm based on pattern recognition，which were applied for the test of the classification precision of prosthesis control system.Six volunteers were enrolled in the test including four kinds of activities and static status through this system.The precision reached to 94.6%，which is better than the accuracy when MMG signal（88.5%）or EMG signal（90.4%）was adopted solo.Conclusions The system based on both of MMG and EMG signals can classify the prosthesis controlling action efficiently，and control the prosthesis independently.This prosthesis control system is expected to be applied to the disabled upper arm amputated in the near future.

参考文献：

［1］Barry DT.Acoustic signals from frog skeletal muscle［J］.Biophys，1987，51：769-773.
［2］Orizio C.Muscle sound： Bases for the introduction of a mechanomyographic signal in muscle studies［J］.Crit Rev Biomed Eng,1993，21：201-243.
［3］Stokes MJ. Acoustic myography： Applications and considerations in measuring muscle performance［J］.Isokinetics and Exercise Science，1993，3：4-15.
［4］Barry DT，Leonard JA，Gitter AJ，et al.Acoustic myography as a control signal for an externally powered prosthesis［J］.Arch Phys Med Rehabil，1986，67：267-269.
［5］Xie HB，Zheng YP，Guo JY.Classification of the mechanomyogram signal using a wavelet packet transform and singular value decomposition for multifunction prosthesis control［J］.Physiological measurement，2009，30（5）：441-457.
［6］Silva J，Heim W，Chau T.A self-contained，mechanomyography-driven externally powered prosthesis［J］.Archives of physical medicine and rehabilitation，2005，86（10）：2066-2070.
［7］Englehart K，Hudgins B，Parker PA，et al.Improving myoelectric signal classification using wavelet packets and principal components analysis［C］.21st Annual International Conference of the IEEE Engineering in Medicine and Biology Society,1999.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录