[1]王力,张雄,仲雪飞,等.时频分析在语言想像脑机接口中的应用[J].东南大学学报(自然科学版),2014,44(6):1126-1130.[doi:10.3969/j.issn.1001-0505.2014.06.006]
 Wang Li,Zhang Xiong,Zhong Xuefei,et al.Application of time-frequency analysis in speech imagery based brain-computer interfaces[J].Journal of Southeast University (Natural Science Edition),2014,44(6):1126-1130.[doi:10.3969/j.issn.1001-0505.2014.06.006]
点击复制

时频分析在语言想像脑机接口中的应用()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
44
期数:
2014年第6期
页码:
1126-1130
栏目:
自动化
出版日期:
2014-11-20

文章信息/Info

Title:
Application of time-frequency analysis in speech imagery based brain-computer interfaces
作者:
王力张雄仲雪飞樊兆雯张玉孙瀚
东南大学电子科学与工程学院, 南京 210096
Author(s):
Wang Li Zhang Xiong Zhong Xuefei Fan Zhaowen Zhang Yu Sun Han
School of Electronic Science and Engineering, Southeast University, Nanjing 210096, China
关键词:
语言想像 脑电信号 时频分析 脑机接口
Keywords:
speech imagery electroencephalography time-frequency analysis brain-computer interfaces
分类号:
TP274
DOI:
10.3969/j.issn.1001-0505.2014.06.006
摘要:
为了分析语言想像所诱导的脑电信号的时间和频率特性, 利用事件相关谱扰动(ERSP)对信号进行时频分析.首先,经ERSP确定能量波动的频率范围;然后,根据该频率范围计算信号的事件相关(去)同步(ERD/ERS), 并利用共空间模式和支持向量机分别对单次实验数据的特征值进行提取和分类.对8位被试的试验结果分析表明, 被试间的频率范围具有显著的差异, 其中4位被试的频率范围含有α波, 3位被试含有α波和β波, 1位被试的脑电信号在默读汉字时无明显变化. 2个汉字默读时的脑电信号可产生相似的ERD/ERS. 优化频率范围后针对这2个汉字的平均分类正确率分别提高了2.25%和1.39%. 时频分析能更好地显示脑电信号的能量变化率, 并能改善语言想像脑机接口的性能.
Abstract:
To analyze time and frequency characteristics of electroencephalography signals induced by speech imagery, event-related spectral perturbation(ERSP)is used in the time-frequency analysis of signals. First, the frequency range of the energy fluctuation can be determined by ERSP. Then, according to the frequency range, event-related(de)synchronization(ERD/ERS)of the signals is calculated, and the eigenvalues of each experimental data are extracted and classified by common spatial pattern and support vector machine, respectively. The results of eight subjects show that the frequency ranges are obviously different among subjects. The ranges of four subjects include α rhythm and those of three subjects include α and β rhythms. The electroencephalography signals of one subject do not evidently change when he reads a Chinese character in mind. The similar ERD/ERS of electroencephalography signals may produce while reading two Chinese characters in mind. After optimizing the frequency ranges, the average classification accuracy for the two characters are improved by 2.25% and 1.39%,respectively. The energy gradient of electroencephalography signals can be exhibited better by time-frequency analysis, and the performance of speech imagery based brain-computer interfaces can be also improved.

参考文献/References:

[1] Yuan H, He B. Brain-computer interfaces using sensorimotor rhythms: current state and future perspectives [J]. IEEE Transactions on Biomedical Engineering, 2014, 61(5): 1425-1435.
[2] McFarland D J, Krusienski D J, Sarnacki W A, et al. Emulation of computer mouse control with a noninvasive brain-computer interface [J]. Journal of Neural Engineering, 2008, 5(2): 101-110.
[3] Hwang H J, Kim S, Choi S, et al. EEG-based brain-computer interfaces: a thorough literature survey [J]. International Journal of Human-Computer Interaction, 2013, 29(12): 814-826.
[4] Friedrich E V C, Scherer R, Neuper C. Stability of event-related(de-)synchronization during brain-computer interface-relevant mental tasks [J]. Clinical Neurophysiology, 2013, 124(1): 61-69.
[5] Pfurtscheller G, Lopes da Silva F H. Event-related EEG/MEG synchronization and desynchronization: basic principles [J]. Clinical Neurophysiology, 1999, 110(11): 1842-1857.
[6] Gouy-Pailler C, Congedo M, Brunner C, et al. Nonstationary brain source separation for multiclass motor imagery [J]. IEEE Transactions on Biomedical Engineering, 2010, 57(2): 469-478.
[7] DaSalla C S, Kambara H, Sato M, et al. Single-trial classification of vowel speech imagery using common spatial patterns [J]. Neural Networks, 2009, 22(9): 1334-1339.
[8] Wang L, Zhang X, Zhong X, et al. Analysis and classification of speech imagery EEG for BCI [J]. Biomedical Signal Processing and Control, 2013, 8(6): 901-908.
[9] Harpaz Y, Lavidor M, Goldstein A. Right semantic modulation of early MEG components during ambiguity resolution [J]. NeuroImage, 2013, 82: 107-114.
[10] Wacker M, Witte H. Time-frequency techniques in biomedical signal analysis [J]. Methods of Information in Medicine, 2013, 52(4): 279-296.
[11] Makeig S. Auditory event-related dynamics of the EEG spectrum and effects of exposure to tones [J]. Electroencephalography and Clinical Neurophysiology, 1993, 86(4): 283-293.
[12] Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis [J]. Journal of Neuroscience Methods, 2004, 134(1): 9-21.
[13] Ramoser H, Muller-Gerking 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.
[14] López Chau A, Li X, Yu W. Convex and concave hulls for classification with support vector machine [J]. Neurocomputing, 2013, 122: 198-209.
[15] Chang C C, Lin C J. LIBSVM—a library for support vector machines [EB/OL].(2014-04-01)[2014-04-08]. http://www.csie.ntu.edu.tw/~cjlin/libsvm.

备注/Memo

备注/Memo:
收稿日期: 2014-04-08.
作者简介: 王力(1986—),男,博士生; 张雄(联系人),男,博士,研究员,博士生导师,zxbell@seu.edu.cn.
基金项目: 国家重点基础研究发展计划(973计划)资助项目(2010CB327705)、国家高技术研究发展计划(863计划)资助项目(2012AA03A302)、中央高校基本科研业务费专项资金资助项目(CXLC12_0095).
引用本文: 王力,张雄,仲雪飞,等.时频分析在语言想像脑机接口中的应用[J].东南大学学报:自然科学版,2014,44(6):1126-1130. [doi:10.3969/j.issn.1001-0505.2014.06.006]
更新日期/Last Update: 2014-11-20