基于共空域频谱模式的少通道运动想象分类
上海交通大学机械与动力工程学院,上海 200240
Common Spatial Spectral Pattern for Motor Imagery Tasks in Small Channel Configuration
School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
摘要 针对少通道脑电数据采集如2~3通道的情况下,利用每个通道信号的2~10个延时来扩展EEG信号的通道数,然后再利用共空域模式CSP算法进行特征抽取。针对每个通道信号的延时选取问题,提出利用非参数化估计互信息熵的方法来选择最佳个数的延时因子。应用多个时延的共空域频谱模式(CSSP)算法,对2008年BCI竞赛数据集IIb中的9个受试者以及实验室采集的13个受试者的想象运动数据集分析,结果表明可以使两个数据集的平均Kappa系数分别达到0.6与0.34。该方法可以依据8~30 Hz内频段的区分度自动优化权重系数,从而提高少通道数目下想象运动的分类正确率。
关键词 :
脑机接口 ,
共空域频谱模式 ,
特征选择 ,
信息熵
Abstract :Targeting to the deficient recording condition with 2~3 channels, a novel method, multiple timedelayed common spatial spectral pattern (CSSP), was proposed in this paper. In this approach, 2~10 timedelayed Electroencephalogram (EEG) signals for each channel were used to expand the number of EEG channels and then common spatial pattern (CSP) was used to extract the features. Mutual information criterion was applied to choose the optimal number of timedelayed signals. The proposed method was tested by the IIb datasets of 9 subjects in BCI Competition 2008 and datasets of 13 subjects in our experiments. The average Kappa coefficients for the two datasets are 0.6 and 0.34, respectively. This method automatically weights the importance of frequency in 8~30 Hz and enhances the classification accuracy in the case of deficient recording consequently.
Key words :
braincomputer interface (BCI)
common spatial spectral pattern (CSSP)
feature selection
mutual information
基金资助: 国家重点基础研究发展计划(973计划)(2011CB013305);上海市科委重点项目(11JC140600);国家重点实验室基金重点项目(MSVZD201204)
[1]Krusienski D, GrosseWentrup M, Galan F, et al. Critical issues in stateofart braincomputer interface signal processing [J]. Journal of Neural Engineering, 2011, 8(2): 025002.
[2]高上凯. 浅谈脑-机接口的发展现状与挑战 [J]. 中国生物医学工程学报, 2007, 26 (6):801-803.
[3]尧德中,刘铁军,雷旭,等. 基于脑电的脑-机接口:关键技术和应用前景 [J]. 电子科技大学学报, 2009, 38(5):550-554.
[4]Pfurtscheller G, Neuper C. Eventrelated synchronization of mu rhythm in the EEG over the cortical hand area in man [J]. Neuroscience Letters, 1994, 174(1):93–96.
[5]Ramoser H, MullerGerking J, Pfurtscheller G. Optimal spatial filtering of single trial EEG during imagined hand movement [J]. IEEE Transactions on Rehabilitation Engineering, 2000, 8(4):441–446.
[6]Lemm S, Blankertz B, Curio G, et al. Spatiospectral filters for improving the classification of single trial EEG [J]. IEEE Transactions on Biomedical Engineering, 2005, 52(9):1541-1548.
[7]Wang Haixian, Tang Qin, Zheng Wenming. L1-normbased common spatial patterns [J]. IEEE Transactions on Biomedical Engineering, 2012, 〖STHZ〗59(3): 653-662.
[8]Lotte F, Guan C. Regularizing common spatial patterns to improve BCI designs: unified theory and new algorithms [J], IEEE Transactions on Biomedical Engineering, 2011, 58(2): 355-362.
[9]Broomhead D, King G. Extracting qualitative dynamics from experimental data [J]. Physica D, 1986, 20(2):217-236.
[10]Fukunaga K. Introduction to Statistical Pattern Recognition [M]. Academic Press, 1990:31-35.
[11]Blankertz B, Tomioka R, Lemm S, et al. Optimizing spatial filters for Robust EEG singletrial analysis [J]. IEEE Signal Processing Magazine, 2008, 25(1):41–56.
[12]Dornhege G, Blankertz B, Krauledat B, et al. Combined optimization of spatial and temproral filters for improving braincomputer interfacing [J]. IEEE Transactions on Biomedical Engineering, 2006, 53(11):2274-2281.
[13]Ang K, Chin Z, Zhang H, et al. Mutual informationbased selection of optimal spatialtemporal patterns for singletrial EEGbased BCIs [J]. Pattern Recognition, 2012, 45(6) : 2137- 2144.
[14]Zhang H, Chin Z, Ang K, et al. Optimum spatiospectral filtering network for braincomputer interface [J]. IEEE Transactions on Neural Networks, 2011, 22(1): 52-63.[15]Duda R, Hart P, Stork D, Pattern classification [M]// Maximumlikelihood and Bayesian Parameter Estimation. New York: John Wiley & Sons, 2001:117-124.
[16]BCI Competition IV. [EB/OL] http:// http://www.bbci.de/competition/iv/desc_2b.pdf, 2008-04-16/2013-01-27.
[17]Meng Jianjun, Huang Gan, Zhang Dingguo, et al. Optimizing spatial spectral patterns jointly with channel configuration for braincomputer interface [J]. Neurocomputing, 2013, 104: 115-126.
[1]
马征 邱天爽* . 视觉ERP脑机接口中实验范式的研究进展 [J]. 中国生物医学工程学报, 2016, 35(1): 96-102.
[2]
曹艳1,2 郑筱祥1,2,3#* . 植入式脑机接口发展概况 [J]. 中国生物医学工程学报, 2014, 33(6): 659-665.
[3]
吴俊1 杨雅2 俞祝良1#* 顾正晖1 李远清1#基于马尔科夫切换过程的运动想象信号分类 [J]. 中国生物医学工程学报, 2014, 33(6): 666-672.
[4]
任玉强1 李丽 杨合龙1 祝磊<. 运用近邻传播聚类分析进行SELDI-TOF蛋白质谱特征选择 [J]. 中国生物医学工程学报, 2013, 32(1): 14-20.
[5]
谢娟英1, 2* 雷金虎1 谢维信2, 3 . 一种新特征评价方法在红斑鳞状皮肤病诊断中的应用 [J]. 中国生物医学工程学报, 2012, 31(1): 59-67.
[6]
王金甲* 胡备. 脑机接口的广义核线性判别分析方法研究 [J]. 中国生物医学工程学报, 2012, 31(1): 75-82.