一种多阈值融合心音递归图的设计与应用

摘要
图/表
参考文献(18)
相关文章 (15)

全文: PDF (965 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为了有针对性地突出心音的某些特征，将心音信号用递归图的方式进行表征和识别。该文首先讨论了心音与递归图的关系，给出适用于心音的递归图获取方法;其次，提出一种自适应阈值计算方法，重点分析如何利用这种自适应阈值构建一种多阈值融合心音递归图，使这种心音递归图能最佳的表现心音特征,并从多阈值融合心音递归图中提取值和灰度共生矩的四种纹理特征：能量、对比度、相关性和熵;最后利用支持向量机进行心音分类识别。实验结果表明，对文中使用的两个数据库，不用进行、分段预处理,平均识别率可达到90%以上，并将这个方法成功应用于心血管健康评估系统。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章

关键词 ：心音, 多阈值融合, 心音递归图, , 灰度共生矩阵

Abstract：To show some characteristics of heart sound, the feature of heart sound signal was characterized by a recurrence plot.Firstly, the relationship between heart sound and a recurrence plot was discussed, then a recurrence plot acquisition method for heart sound was given.Secondly, an adaptive threshold calculation method was proposed, then a multi-threshold fusion recurrence plot of heart sound was constructed, so that the recurrence plot of heart sound could best present the characteristics of heart sound.Next, four kinds of texture features of gray-level co-occurrence-matrix and the value of D′2/S′2 were extracted from the multi-threshold fusion recurrence plot of heart sound.Finally support vector machine was used to realize identification.The experimental results show that the average accuracy could reach to 95% without performing of S1，S2 segmentation.The method had been successfully applied to a cardiovascular health assessment system.

Key words： heart sound multi-threshold fusion recurrence plot of heart sound ;Gray-Level Co-occurrence-Matrix(GLCM)

收稿日期: 2017-11-21 出版日期: 2019-08-15

引用本文:

成谢锋1,2，汪晶1,王悦1. 一种多阈值融合心音递归图的设计与应用[J]. 振动与冲击, 2019, 38(16): 108-114.
CHENG Xiefeng1,2 WANG Jing1 WANG Yue1. Design and application of a multi-threshold fusion recurrence plot of heart sound. JOURNAL OF VIBRATION AND SHOCK, 2019, 38(16): 108-114.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2019/V38/I16/108

[1]成谢锋,马勇,刘陈,张学军,郭宇锋.心音身份识别技术的研究[J].中国科学:信息科学,2012,42(02):235-249.
CHENG Xie-feng, MA Yong, LIU Chen, ZHANG Xue-jun ,and GUO Yu-feng. An introduction to heart sounds identification technology[J].Science China: Information Science,2012,42(02):235-249.
[2]Eckmann J P, Kamphorst S O, Ruelle D.Recurrence plots of dynamical systems[J]. EPL (Europhysics Letters), 1987, 4(9): 973.
[3]Mindlin G M,Gilmore R..Topological analysis and synthesis of chaotic time series[J]. Physica D: Nonlinear Phenomena, 1992, 58(1-4): 229-242.
[4]Schlenker J, Socha V, Riedlbauchová L, et al. Recurrence plot of heart rate variability signal in patients with vasovagal syncopes[J]. Biomedical Signal Processing and Control, 2016, 25: 1-11.
[5]Zbilut J P, Zaldivar-Comenges J M, Strozzi F. Recurrence quantification based liapunov exponents for monitoring divergence in experimental data[J]. Physics Letters A, 2002, 297(3): 173-181.
[6]丁晓蓉,郭兴明,钟丽莎.基于混沌理论的心音信号非线性动力学分析[J].振动与冲击.2012,31(15):55-59.
DING Xiao-rong, GUO Xing-ming, ZHONG Li-sha. Nonlinear dynamic analysis of heart sound signals based on chaos theory[J].Journal of Vibration and Shock. 2012,31(15):55-59.
[7] 李慧娜, 郭超峰, 平源.灰度共生矩阵在指纹图像分割中的应用[J]. 数据采集与处理, 2012, 27(1):63-67.
LI Hui-na, GUO Chao-feng, PINF Yuan. Application of gray level co-occurrence matrix for fingerprint segmentation[J].Journal of Data Acquisition & Processing. 2012, 27(1):63-67.
[8]张恩瑜,师永民,张国山等.基于交叉验证法探究砾岩储集层特征及对化学驱油剂损耗的影响[J].科学技术与工程,2017,(12):45-54.
ZHANG En-yu, SHI Yong-ming, ZHANG Guo-shan ,et al. Study on conglomerate reservoir characteristics and their influenceson oil displacement agent loss based on cross validation[J]. Science Technology and Engineering. 2017,(12):45-54.
[9]成谢锋,严誌,马勇,等.运动与年龄对心音混沌特性影响规律的研究[J]. 振动与冲击, 2017, 36(1):175-180.
CHENG Xie-feng, YAN Zhi, Ma Yong,et al. Influences of exercise and age on heart sound chaotic characteristics[J]. Journal of Vibration and Shock. 2017, 36(1):175-180.
[10]吴庆涛,曹再辉,施进发.基于改进颜色直方图和灰度共生矩阵的图像检索[J].图学学报,2017,(04):543-548.
WU Qing-tao, CAO Zai-hui , SHI Jin-fa. Image retrieval based on improved color histogram and gray level co-occurrence matrix[J]. Journal of Graphics 2017,(04):543-548.
[11] CLIFFORD G D, LIU C Y, MOODY B E，et al. Recent advances in heart sound analysis[J]. Physiological Measurement, 2017, 38(8):10-25.
[12]郭兴明,李传鹏,卢德林.定量递归分析和近似熵在心音分类识别中的应用[J].数据采集与处理,2013,28(05):559-564.
GUO Xing-ming, LI Chuan-peng , LU De-lin. Application of recurrence quantification analysis and approximate entropy in classification and recognition of heart sound[J].Journal of Data Acquisition and Processing,2013,28(05):559-564.
[13]李新杰,胡铁松,董秀明.递归图法在径流时间序列非线性分析中的应用[J].武汉大学学报,2013,46(01):62-66+72.
LI Xin-jie, HU Tie-song, DONG Xiu-ming. Application of recurrence plots to nonlinear analysis of runoff time series[J].Engineering Journal of Wuhan University,2013,46(01):62-66+72.
[14]Schinkel S, Dimigen O, Marwan N. Selection of recurrence threshold for signal detection[J]. The European Physical Journal-Special Topics, 2008, 164(1): 45-53.
[15]Thiel M, Romano M C, Kurths J, et al. Influence of observational noise on the recurrence quantification analysis[J]. Physica D: Nonlinear Phenomena, 2002, 171(3): 138-152.
[16]Kocyigit Y , Seker H. Imbalanced data classifier by using ensemble fuzzyc-means clustering[C]//Biomedical and Health Informatics (BHI), 2012 IEEE-EMBS International Conference on. IEEE, 2012: 952-955.
[17]成谢锋,蔡华民.二维心音图特征提取与识别方法的研究[J].图学学报,2014,35(02):268-273.
CHENG Xie-feng, CAI Hua-ming. Research on methods of feature extraction and recognition of two-dimensional phonocardiogram[J].Journal.of.Graphics,2014,35(02):268-273.
[18]YANG H. Multiscale recurrence quantification analysis of spatial cardiac vectorcardiogram signals[J]. IEEE Transactions on Biomedical Engineering, 2011, 58(2): 339-347.