基于经验模式分解心音自动分段算法.pdfVIP

27 卷 4 期 中　国　生 　物 　医 　学 　工 　程 　学 　报 Vol . 27 　No . 4 2008 年 8 月 Chinese J ournal of B iomedical Engineering August 　2008 基于经验模式分解的心音自动分段算法 　 林 　勇 　许晓飞 ( 山东大学控制科学与工程学院生物医学工程研究所 ,济南 　250061) 摘 　要 : 与原始信号相比 ,心音包络能够更好地显示心音的特征 。心音包络的提取是对心音进行时域分析的基 础 。本研究提出利用希尔伯特黄变换提取心音包络的方法 。首先利用黄变换提取心音信号的固有模态函数 ,然 后利用希尔伯特变换提取心音包络 。在希尔伯特黄变换提取的心音包络基础上 , 对 50 例正常人心音样本进行了 第一心音 、第二心音识别 ,正确率达到了92 % ,为下一步的心音分析及诊断奠定了基础 。 关键词 : 心音 ; 包络提取 ; 经验模式分解 Segmentation Algorithm of Heart Sounds Based on Empirical Mode Decomposition LIN Yong 　XU XiaoFei ( Institute of B iomedical Engineering , School of Control Science and Engineering , Shandong University , J inan 250061) Abstract : Envelope of heart sounds can reflect the features of heart sounds better than original heart sounds do . Envelope extraction of heart sounds is the foundation for time domain analysis. In this paper , a new method was proposed for envelope extraction of heart sounds based on HilbertHuang transform. Firstly , in order to extract intrinsic mode functions , the original heart sounds signal was preprocessed by Huang transform. Secondly , the envelope of heart sounds was extracted by Hilbert transform. And based on the envelope extracted , the first and second heart sounds of 50 samples from normal persons were recognized with the accuracy of 92 %. Key words :heart sounds ; envelope extraction ; empirical mode decomposition ( ) 中图分类号 　R31804 　　文献标识码 　A 　　文章编号 2008