基于虚拟通道的ICA的P-VEP提取方法的研究  

A Method of the Estimation of Pattern Visual Evoked Potential (P-VEP) Based on the ICA of Virtual Channels

作　　者：刘钦团[1] 邱飞岳[1,2] 李浩君[2] 

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310032 [2]浙江工业大学教科学院,浙江杭州310012

出　　处：《计算机工程与科学》2010年第7期137-139,共3页Computer Engineering & Science

基　　金：浙江省科技计划资助项目(2006C31001)

摘　　要：本文介绍了独立分量分析算法(ICA)的原理、算法及其特点。采用ICA算法进行P-VEP的提取,考虑到仪器实际应用中单通道VEP信号的提取,本文通过引入虚拟通道,构造虚拟多通道,克服了ICA无法直接对单路信号进行独立分量分析的难题,实现P-VEP单通道单次的有效提取。经与目前临床通用的相干平均法比较,只需单通道、单次,在波形整体和P100潜伏期的提取上,效果显著,获得医师欢迎,很有进一步开发潜力。The principle and algorithms of independent component analysis（ICA） is illustrated. The paper introduces virtual channels and the virtual structure of multichannel, to complete a singletrial estimation of singlechannel pattern PVEP, which overcomes the problem of using ICA directly to singlechannel signals. Compared with the average technology, its encouraging results are illustrated by both computer simulation and clinical data applications. The number of trials needed is reduced to one. By analyzing the time course and spatial pattern of the independent components, a certain component is found to be closely related with the P100 latent period in the VEP complex.

关 键 词：ICA P-VEP 虚拟通道 单次提取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]