文献类型：期刊文献

中文题名：基于FastICA的P300电位快速提取方法

英文题名：A FastICA-based Approach to Extracting P300 Potential

作者：张宇[1];张建华[1];王行愚[1];金晶[1]

机构：[1]华东理工大学信息科学与工程学院,上海200237

年份：2009

卷号：35

期号：5

起止页码：750

中文期刊名：华东理工大学学报（自然科学版）

外文期刊名：Journal of East China University of Science and Technology

收录：CSTPCD;;Scopus;北大核心:【北大核心2008】；CSCD:【CSCD2011_2012】；

基金：国家自然科学基金项目(60775033;60674089);上海市浦江人才计划项目(07PJ14031);上海市重点学科项目(B504)

语种：中文

中文关键词：快速独立分量分析;脑电信号EEG;靶刺激;残疾被试

外文关键词：FastICA; electroencephalogram （EEG）; target stimuli; disabled subjects

摘要：从两个方面研究了快速独立分量分析(Fast Independent Component Analysis,FastICA)方法在诱发脑电P300少次提取中的应用,并给出了针对健康和残疾被试的实验结果。首先,利用FastICA对观测信号进行去噪,然后对去噪后的P300分量进行较少次叠加平均,并对提取出的健康和残疾被试的P300特征进行了详细的比较分析;然后,从模式识别的角度出发,逐渐减少叠加次数,分别考察了根据提取出的P300特征进行靶刺激和非靶刺激识别的难易程度。实验结果表明了FastICA方法用于P300较少次提取的有效性。
From two aspects, this paper analyzes the application of fast independent component analysis （FastlCA） algorithm in the extraction of P300 by a few trials averaging, and presents simulation results for both normal and disabled subjects. Firstly, the noises are removed from the observed signals by using FastICA, and a few P300 trials are averaged. The features of P300 for both normal and disabled subjects are analyzed. Secondly, in terms of pattern recognition, the number of trials is reduced successively, and the difficulties of distinguishing targets stimuli from non-target ones based on the features are analyzed. The results illustrate the validity of FastlCA in the application of the fast extraction of P300.