膝关节声发射信号的统计分析与模式识别  

Statistical analysis and pattern recognition of knee joint acoustic emission signals

作　　者：张洪波[1] 庞月强 李小亭[1] 王博 Zhang Hongbo;Pang Yueqiang;Li Xiaoting;Wang Bo(College of Quality and Technical Supervision,National&Local Joint Engineering Research Center of Metrology Instrument and System,Hebei Key Laboratory of Energy Metering and Safety Testing Technology,Hebei University,Baoding 071002,China)

机构地区：[1]河北大学质量技术监督学院,计量仪器与系统国家地方联合工程研究中心,河北省能源计量与安全检测技术重点实验室,保定071002

出　　处：《电子测量与仪器学报》2021年第8期198-204,共7页Journal of Electronic Measurement and Instrumentation

摘　　要：为了实现膝关节声发射信号的动态分析和模式识别,以膝关节在坐-立-坐运动的不同阶段产生的声发射信号为研究对象,进行主成分分析、差异性检验和基于支持向量机的分类测试。声发射特征参数经过线性变化提取为2个主成分;对膝关节在两个运动阶段产生的声发射信号进行差异性检验,渐进显著性健康组主成分F_(1)<0.05、主成分F_(2)>0.05,对照组均小于0.05;支持向量机对膝关节声发射信号的分类准确率达到了97.9%。结果表明,主成分分析的方法能够对膝关节声发射信号成功降维,对降维后的信号进行差异性检验和诊断识别发现患病膝关节的不同运动阶段存在更为明显的差异,支持向量机的方法能够对膝关节骨性关节炎做出准确的诊断识别。In order to realize dynamic analysis and pattern recognition of knee joint acoustic emission signals,principal component analysis,difference test and classification test based on support vector machine were carried out to study the acoustic emission signals generated by knee joints in different stages of sitting-standing-sitting.The characteristic parameters of the acoustic emission signals were extracted into two principal components after linear changes;the difference test of the acoustic emission signals generated by the knee joint in the two motion stages shows that the result of the progressive significance of the healthy group was the principal component F_(1)<0.05,the principal component F_(2)>0.05,the progressive significance results of the control group was less than 0.05;The classification accuracy of the support vector machine for the acoustic emission signal of the knee joint reached 97.9%.The results show that principal component analysis can successfully reduce the dimension of knee acoustic emission signals;the acoustic emission signals at different stages of movement are different,which is particularly obvious in the diseased knee joint;the support vector machine method can accurately diagnose and identify.

关 键 词：膝关节 声发射 主成分分析 支持向量机 

分 类 号：TN06[电子电信—物理电子学]