经验模式分解及关联维数在心音信号分类识别中的应用.pdfVIP

第42 卷 第6 期 电 子 科 技 大 学 学 报 Vol.42 No.6 2013年11月 Journal of University of Electronic Science and Technology of China Nov . 2013 经验模式分解及关联维数在心音信号分类识别中的应用 郭兴明，袁志会，丁晓蓉 (重庆大学生物工程学院 重庆 沙坪坝区 400044) 【摘要】针对心音信号非线性、非平稳的特性，提出一种基于经验模式分解(EMD)和关联维数的心音特征提取方法。首 先通过EMD方法将心音信号分解成若干个固有模态函数(IMF) ，并利用互相关系数准则对IMF进行筛选，结合G-P算法对主 IMF(IMF1~IMF4)分量分别求其关联维数，以此作为神经网络的输入向量，实现了对正常心音信号和病理心音信号的分类识别。 对于重构相空间中的两个重要参数时间延迟τ和关联维数m ，分别采用互信息函数法和用Cao算法确定。对临床采集的心音数 据按该方法进行测试，结果表明，该方法能有效地识别心音。 关 键 词 关联维数; 经验模式分解; 心音; 神经网络 中图分类号 TP274 文献标志码 A doi:10.3969/j.issn.1001-0548.2013.06.028 Application of EMD and Correlation Dimension in Classification and Recognition of Heart Sound GUO Xing-ming, YUAN Zhi-hui, and DING Xiao-rong (College of Bioengineering, Chongqing University Shapingba Chongqing 400044) Abstract Focusing on the non-stationary and non-linear of heart sounds, a new method of feature extraction based on empirical mode decomposition (EMD) and Correlation dimension is proposed. The heart sound signals are decomposed into a finite number of intrinsic mode functions (IMFs). The IMF components are chosen by using the criteria of mutual correlation coefficient between IMF components and original signal and then the correlation dimension of the top four intrinsic mode functions (IMF1~IMF4) is calculated by using G-P algorithm. The eigenvectors are put into the artificial neural network for automatic discrimination between normal and abnormal signals. In the process of phase-space reconstruction, Cao theory and the mutual information function are used to determine the two important parameters: delay time and emb