期刊文献+

基于sEMG与K-means聚类的上肢痉挛状态定量评定方法 被引量:6

Quantitative evaluation method of upper limb spasticity based on s EMG signals and K-means clustering algorithm
下载PDF
导出
摘要 为解决临床痉挛状态评定主观性大,信度效度待提高的问题,提出基于表面肌电(surface electromyography,s EMG)信号与K-means聚类的上肢痉挛状态定量评定方法。结合时频域分析与非线性动力学分析方法,利用s EMG信号希尔伯特-黄变换(Hilbert-Huang transform,HHT)边际谱熵判定牵张反射肌电阈值(stretch reflex onset,SRO);取SRO后固定长度s EMG信号均方根(root mean square,RMS)个体差异消除后得到的均方根差值(RMS difference,RMSD)评定痉挛状态,利用K-means聚类算法对RMSD分类以重新评估痉挛状态。实验数据表明HHT边际谱熵能够准确识别SRO(识别率:95%),且在背景含尖锐毛刺噪声以及短时信号处理上表现出很好的性能;RMSD与改良Ashworth量表(modified Ashworth scale,MAS)评分显著相关(test:r=0.843,r2=0.711,p<0.01;retest:r=0.836,r2=0.699,p<0.01),重测信度良好;基于K-means聚类算法的痉挛状态等级与RMSD相关性为(test:r=0.946,r2=0.895,p<0.01;retest:r=0.942,r2=0.887,p<0.01),且各组之间差异性显著(p<0.01)。实验结果表明,该方法可为上肢痉挛状态评定提供一种客观定量的分析手段,相比于MAS能更好的定量评定与细分痉挛状态。 In order to assess spasticity level quantitatively and improve assessment validity and reliability,a method based on surface electromyo graphy( s EMG) signals and K-means clustering algorithm is proposed. Combining nonlinear dynamics and time-frequency analysis,the Hilbert-Huang transform( HHT) marginal spectrum entropy of s EMG signals was employed for stretch the reflex onset( SRO) detection and the root mean square difference( RMSD),which is the difference between the RMS of fixed-length s EMG after the SRO and the RMS of base line s EMG,was used to evaluate spasticity. The spasticity level was reevaluated based on K-means clustering of the RMSD. The experimental results show that the HHT marginal spectrum entropy can precisely detect SRO and shows strong robust in processing short length time data series and against spurious background spikes. There was a strong correlation( test: r = 0. 843,r^2=0. 711,p < 0. 01; retest: r = 0. 836,r^2= 0. 699,p < 0. 01) between spasticity level and RMSD. Also,there was good test-retest reliability. Besides,our groups classified by the criteria of the RMSD had a strong correlation( test: r = 0. 946,r^2= 0. 895,p < 0. 01;retest: r = 0. 942,r^2= 0. 887,p < 0. 01) with the RMSDs,and there were significant differences between the RMSDs of each group( p < 0. 01). The experimental results demonstrate that this method can provide an objective and quantitative way for spasticity measurement and may be an alternative clinical measure to the modified Ashworth scale( MAS).
出处 《电子测量与仪器学报》 CSCD 北大核心 2018年第6期53-63,共11页 Journal of Electronic Measurement and Instrumentation
基金 国家自然科学基金(51279044) 市级科技计划重点项目(2015cy04)资助
关键词 痉挛状态评定 表面肌电信号 HHT边际谱熵 均方根 K-MEANS聚类算法 spasticity assessment sEMG HHT marginal spectrum entropy root mean square K-means clustering algorithm
  • 相关文献

参考文献11

二级参考文献132

共引文献296

同被引文献52

引证文献6

二级引证文献19

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部