WICA算法在脑电P300波成分提取中的应用研究  

P300 Wave Extraction by Wavelet Transform Combined with Independent Component Analysis

作　　者：傅伟[1] 张新友[2] 熊平[3] 

机构地区：[1]江苏财经职业技术学院计算机工程系,江苏淮安223003 [2]湖南师范大学数学与计算机科学学院,长沙410081 [3]中南大学信息物理工程学院,长沙410083

出　　处：《医疗卫生装备》2009年第1期4-6,9,共4页Chinese Medical Equipment Journal

基　　金：国家自然科学基金项目(30371626)

摘　　要：目的:研究新的脑电波成分提取方法,以便更有效地提取脑电P300波成分。方法:根据脑电信号以及信号分析方法各自的特点,把小波变换(WT)和独立分量(ICA)相结合的方法(WICA)应用到脑电P300波成分提取中。结果:小波变换可以增强待检测信号成分,削弱非目标信号成分和噪声的干扰;在经过小波变换后子带重组的多道ICA输入信号中,非目标信号成分和干扰信号成分已变得较弱,因此WICA算法能有效地分离出相对较强的目标信号成分。结论:实验结果表明,WICA算法在脑电P300波成分提取中取得了比较理想的效果。Objective To study P300 wave extraction by a new method. Methods According to the respective characteristics of the brain electrical signal and the signal analysis method, the algorithm by Wavelet Transformation （WT） combined with Independent Component Analysis （ICA）, abbreviated to WICA, was applied to P300 wave extraction. Results With WT, the target component of the signal could be strengthened, while the non-target component and the noise were weakened which became weak in multi-channel ICA input signal, and then WICA algorithm could effectively separate strong target component. Conclusion WICA algorithm behaves well in P300 wave extraction.

关 键 词：小波变换 独立分量分析 脑信号 特征增强 

分 类 号：R318.0[医药卫生—生物医学工程]