电子学报

• 科研通信 • 上一篇    下一篇

基于ICA的脑电信号P300少次自动提取

陈洪波1, 李蓓蕾1,2, 陈真诚1   

  1. 1. 桂林电子科技大学生命与环境科学学院,广西桂林 541004;
    2. 桂林电子科技大学电子工程与自动化学院,广西桂林 541004
  • 收稿日期:2011-06-28 修回日期:2011-12-21 出版日期:2012-06-25
    • 作者简介:
    • 陈洪波 男,博士,副教授,1972年出生于湖南株洲.桂林电子科技大学生命与环境科学学院副院长,主要从事生物医学信息处理方面研究. E-mail:hongbochen@163.com
      李蓓蕾 女,硕士研究生,1986年出生于河北邢台.研究方向为生物医学信号检测与信息处理. E-mail:beilei040810@163.com
      陈真诚 男,博士,教授,博士生导师,1965年出生于湖南永州.桂林电子科技大学生命与环境科学学院院长,主要从事生物传感器与智能仪器、生物医学信息处理方面的研究. E-mail:chenzhcheng@163.com
    • 基金资助:
    • 国家863高技术研究发展计划 (No.2006AA06Z105); 广西教育厅科研项目 (No.201012MA084)

Automatically Extract P300 Within Several Trials from EEG Based on ICA

CHEN Hong-bo1, LI Bei-lei1,2, CHEN Zhen-cheng1   

  1. 1. School of Life and Environmental Science,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China;
    2. School of Electronic Engineering and Automatic,Guilin University of Electronic Technology,Guilin,Guangxi 541004,China
  • Received:2011-06-28 Revised:2011-12-21 Online:2012-06-25 Published:2012-06-25
    • Supported by:
    • National 863 High-tech R&D Program of China (No.2006AA06Z105); Research Program of Education Department of Guangxi Zhuang Autonomous Region (No.201012MA084)

摘要: 提出一种基于Infomax ICA少次自动提取脑电信号P300成分的方法.为了提高ICA分解的有效性,对原始数据中的自发脑电信号和P300成分进行了均衡.混合信号经过ICA分解后,根据IC的固定时间模式的标准差来自动选择P300成分IC,最后重构得到P300成分.实验结果是:利用6试次实验数据经过本文方法处理后能自动得到P300成分,与29试次平均结果(标准信号)相比,它们之间的Pearson相关系数达0.9035,而6试次实验数据平均的结果与标准信号之间的Pearson相关系数为0.5105.结果表明,该方法能有效的获取P300成分,同时增强了P300成分少次提取的客观性.

关键词: 独立分量分析, P300, 脑电, 固定时间模式

Abstract: This paper puts forward a method for automatically extracting the P300 from electroencephalography (EEG) signals within several trials based on Infomax independent component analysis (ICA).An algorithm for signaling equilibrium is proposed to enhance the effectiveness of ICA decomposition.After the mixed signal is decomposed by Infomax ICA,the independent component (IC) of P300 is automatically selected according to the standard deviation of the fixed-temporal-pattern of the IC,and applied in P300 reconstruction.Experimental results show that the P300 can be obtained automatically after six trials on the experimental data,and the result of its Pearson correlation coefficient (PCC) within the average of 29 trials (standard signal) is 0.9035.However,the PCC of the average result of six trials and standard signal is only 0.5105,demonstrating the practical applicability of Infomax ICA.This algorithm enhances the objectivity of P300 extraction within several trials.

Key words: independent component analysis(ICA), P300, electroencephalography(EEG), fixed-temporal-pattern

中图分类号: