中国组织工程研究

中国组织工程研究 ›› 2011, Vol. 15 ›› Issue (13): 2394-2397.doi: 10.3969/j.issn.1673-8225.2011.13.027

• 数字化骨科 digital orthopedics • 上一篇    下一篇

基于独立成分和小波分析的胎儿心电提取

杜艳琴，黄  华   

  1. 四川大学电气信息学院医学信息工程，四川省成都市  610065
  • 收稿日期:2010-10-09 修回日期:2011-01-05 出版日期:2011-03-26 发布日期:2013-10-23
  • 作者简介:杜艳琴★，女，1987年生，四川省南充市人，汉族，四川大学在读硕士，主要从事医学仪器与医学信号的研究。 duyanqin@126.com

Fetal electrocardiogram extraction based on independent component and wavelet analyses

Du Yan-qin, Huang Hua   

  1. School of Electrical Engineering and Information, Sichuan University, Chengdu  610065, Sichuan Province, China
  • Received:2010-10-09 Revised:2011-01-05 Online:2011-03-26 Published:2013-10-23
  • About author:Du Yan-qin★, Studying for master’s degree, School of Electrical Engineering and Information, Sichuan University, Chengdu 610065, Sichuan Province, China duyanqin@126.com

PDF

554

被引次数

12

摘要：

背景：在胎儿心电信号的采集过程中，会受到母体和其他噪声的强干扰，如何快捷与有效地提取出胎儿心电将成为重要的研究课题。
目的：采用结合独立成分分析和小波分析的方法对来自于同一母体的观测信号进行独立分量分离，得到有效的胎儿心电。
方法：结合独立成分分析和小波分析的算法进行胎儿心电的特征提取，首先对含噪信号进行小波变换，去除奇异信号和非平稳随机信号，然后对小波重构后的信号运用快速独立成分分析算法进行成分分析。
结果与结论：在胎儿心电信号的采集过程中，会受到母体和其他噪声的强干扰，但这些信号都是随机的，不相关的，可以认为它们间是相互独立的。采用结合独立成分和小波分析的方法对来自于同一母体的观测信号进行独立分量分离，得到有效的胎儿心电。实验证明该方法是一种有效的方法。

关键词: 独立成分分析, 小波分析, 胎儿, 心电, 信号

Abstract:

BACKGROUND: In the fetal electrocardiogram (ECG) acquisition process, it is interfered by maternal ECG and other noise. It becomes important that how to extract rapidly and effectively fetal ECG.
OBJECTIVE: Using independent component analysis and wavelet analysis to isolate maternal detection signals to obtain effective fetal ECG.
METHODS: Fetal ECG was extracted using the method combining independent component analysis and wavelet analysis. First, noised signals underwent wavelet transform to eliminate strange signals and non-stable random signals, followed by rapid independent component analysis.
RESULTS AND CONCLUSION: Fetal ECG collection was interfered by maternal ECG and other noise, but these signals are random, not related, and can be considered independent of each other. Each independent component could be extracted effectively by using the method combining independent component analysis and wavelet analysis. Experiments show that the method is effective.

中图分类号: