基于离散平稳小波和独立分量分析的事件相关电位单次提取方法

作者：邱红,余世明,张瑞 日期：2009-02-23/span> 浏览：4068 查看PDF文档

基于离散平稳小波和独立分量分析的事件相关电位单次提取方法

邱红,余世明,张瑞
（浙江工业大学 信息学院，浙江 杭州 310014）

摘要：如何在极低的信噪比下快速提取特征信号是事件相关电位（ERP）快速提取的关键技术。提出了将离散平稳小波和独立分量分析（ICA）相结合的方法以去除噪声，并提取事件相关电位。采用离散平稳小波变换分解ERP，选取多个尺度的子带信号，滤除高频噪声对应的小波系数；将串接小波系数作为独立分量分析的输入，利用FastICA算法实现事件相关电位的快速提取。仿真实验结果表明，与传统的相关平均法相比，该方法获得的结果较为满意；与单独采用独立分量分析方法相比，该方法的收敛速度更快。
关键词：离散小波；独立分量分析；事件相关电位；单次提取
中图分类号：TP391; R318文献标识码：A文章编号：1001-4551(2009)02-0088-03

Singletrial extraction of eventrelated potentials based on discrete stationary
wavelet and independent component analysis
QIU Hong, YU Shiming, ZHANG Rui
(College of Information Engineering, Zhejiang University of Technology, Hangzhou 310014, China)
Abstract: It is very difficult to extract eventrelated potentials(ERP) from spontaneous rhythms under quite a low S/N ratio. Discrete stationary wavelet transformation combined with independent component analysis （ICA） was proposed to remove noises and extract eventrelated potential. Firstly, ERP was decomposed by discrete stationary wavelet transformation, multiscale signals were selected. Wavelet coefficients that combined with highfrequency noises were removed. Then all wavelet coefficients were connected to perform ICA. The simulation results show that, compared with traditional averaging method, the method can simplify extraction. It also achieves a faster convergence than simple ICA method.
Key words: discrete transform; independent component analysis (ICA); eventrelated potentials (ERP); singletrial extraction
参考文献(References):
［1］REGAN D. Human Brain Electrophysiology: Evoked Potentials and Evoked Magnetic Field in Science and Medicine ［M］. New York: Elsevier Science Publishing, 1989.
［2］王荣昌，都思丹.基于参数模型和独立分量分析的事件相关诱发电位单次提取［J］.生物医学工程学杂志,2006,23(6):1222-1227.
［3］刘洋，邱天爽，毕晓辉.基于独立分量分析和遗传算法的诱发电位提取新方法［J］.中国生物医学工程学报,2007,26(3):349-354.
［4］SEYEDEHMINA L S. Wavelet Filtering of the P300 Component in EventRelated Potentials ［C］. Proceedings of the 28th IEEE EMBS Annual International Conference. New York:［s.n］, 2006.
［5］李章勇，赵志强，刘圣蓉，等.基于多分辨分析与连续小波变换提取和分析兔体感诱发电位［J］.生物医学工程学杂志，2007，24(3)：504-508.
［6］HYVARINON A,IHA E. A fast fixed|point algorithm for independent component analysis［J］. Neural Computation,1997,9(7):1483-1492.
［7］GAO Xiaorong, XU Neng. Optimal Selection of Independent Components for Eventrelated Brain Electrical Potential Enhancement［C］//2004 IEEE International Workshop on Biomedical Circuits & Systems. Singqpore:［s.n］,2004:S3/5/INV-S3/5/1-4.
［8］叶红仙，项文娟，刘仲宇，等.基于小波包ICA的齿轮箱信号分离系统［J］.机电工程，2008,25(4):5-7.