Please wait a minute...
浙江大学学报(工学版)  2017, Vol. 51 Issue (7): 1381-1389    DOI: 10.3785/j.issn.1008-973X.2017.07.016
自动化技术     
基于卷积神经网络的脑电信号上肢运动意图识别
王卫星1,2, 孙守迁1, 李超1, 唐智川1
1. 浙江大学 计算机科学与技术学院, 浙江 杭州 310007;
2. 贵州大学 机械工程学院, 贵州 贵阳 550025
Recognition of upper limb motion intention of EEG signal based on convolutional neural network
WANG Wei-xing1,2, SUN Shou-qian1, LI Chao1, TANG Zhi-chuan1
1. College of Computer Science and Technology, Zhejiang University, Hangzhou 310007, China;
2. College of Mechanical Engineering, Guizhou University, Guiyang 550025, China
 全文: PDF(2831 KB)   HTML
摘要:

为了在脑机交互中能够对运动意图进行识别,使设备能够预判人的行为动作并提前作出反应,脑电(EEG)信号运用学习过程去解码,并建立识别机制.针对传统生物信号模式识别模型中手动提取特征可能会产生信息损失的问题,引入深度学习的卷积神经网络(CNN),并和目前广泛使用的两种特征提取方法使用BP神经网络分类进行对比.结果显示,CNN在左、右手2分类动作和单手3分类动作中,提高识别精度分别约为4%和8%,增加了动作预测的可靠性.通过对上肢运动意图识别的讨论,可以更好地进行脑机交互控制,并加深对中枢神经信号与手部动作关系的理解.

Abstract:

Electroencephalogram (EEG) signal was decoded using the learning process and the recognition mechanism was established in order to identify the motion intention in the brain-computer interaction and make device predict the behavior of the person and respond in advance. In the traditional biological signal pattern recognition model, the manual extraction of features may cause the problem of information loss. Convoluted neural network (CNN) of depth learning was introduced to identify the motion intention based on the EEG signal and compared with the two feature extraction methods widely used at present that use BP neural network. Results showed that CNN improved the recognition accuracy about 4% and 8% in the experiments of left-right hand 2 classification and one-hand 3 classification, and increased the reliability of the prediction. The brain-computer interaction control can be better conducted through the discussion of the upper limb motor intention recognition. The understanding of the relationship between central nervous signals and hand motions was further enhanced.

收稿日期: 2017-01-02 出版日期: 2017-07-08
CLC:  TP241  
基金资助:

国家自然科学基金资助项目(61303137,61402141);国家教育部博士点基金资助项目(20130101110148)

通讯作者: 孙守迁,男,教授.ORCID:0000-0001-7204-6188.     E-mail: ssq@zju.edu.cn
作者简介: 王卫星(1982—),男,博士生,讲师,从事数字化设计、人机交互的研究.ORCID:0000-0001-7847-0019.E-mail:wwx515@sina.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  

引用本文:

王卫星, 孙守迁, 李超, 唐智川. 基于卷积神经网络的脑电信号上肢运动意图识别[J]. 浙江大学学报(工学版), 2017, 51(7): 1381-1389.

WANG Wei-xing, SUN Shou-qian, LI Chao, TANG Zhi-chuan. Recognition of upper limb motion intention of EEG signal based on convolutional neural network. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2017, 51(7): 1381-1389.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2017.07.016        http://www.zjujournals.com/eng/CN/Y2017/V51/I7/1381

[1] ANG K K, CHUA K S G, PHUA K S, et al. A randomized controlled trial of EEG-based motor imagery brain-computer interface robotic rehabilitation for stroke [J]. Clinical EEG and Neuroscience, 2015, 46(4):310-320.
[2] KIGUCHI K, HAYASHI Y. Motion estimation based on EMG and EEG signals to control wearable robots [C]//IEEE International Conference on Systems, Man and Cybernetics (SMC). [S. l.]: IEEE, 2013:4213-4218.
[3] PFURTSCHELLER G, NEUPER C. Movement and ERD/ERS [M]. The Bereitschaftspotential, US: Springer, 2003: 191-206.
[4] LEW E, CHAVARRIAGA R, SILVONI S, et al.Detection of self-paced reaching movement intention from EEG signals [J]. Frontiers in Neuroengineering, 2012,5(13): 13-18.
[5] KORIK A, SOSNIK R, SIDDIQUE N, et al. 3D hand motion trajectory prediction from EEG mu and beta bandpower [J]. Progress in Brain Research, 2016,22(8): 71-105.
[6] PLOW E B, ARORA P, PLINE M A, et al. Within-limb somatotopy in primary motor cortex-revealed using fMRI [J]. Cortex, 2009, 46(3): 310-321.
[7] VUCKOVIC A, SEPULVEDA F. Delta band contribution in cue based single trial classification of real and imaginary wrist movements [J]. Medical Biological Engineering Computing, 2008, 46(6): 529-539.
[8] GHANI F, SULTAN H, ANWAR D, et al. Classification of wrist movements using EEG signals [J]. Journal of Next Generation Information Technology, 2013, 4(8): 29-39.
[9] KRAHENBUHL P, DOERSCH C, DONAHUE J, et al. Data-dependent initializations of convolutional neural networks [J]. Computer Science, 2015, 11(1): 1-12.
[10] 曾祥炎,陈军.E-Prime实验设计技术[M].广州:暨南大学出版社,2009: 45-47.
[11] EDELMAN B J, BSXTER B, HE B. EEG Source imaging enhances the decoding of complex right hand motor imagery tasks [J]. IEEE Transactions on Biomedical Engineering, 2015, 63(1): 4-14.
[12] FORMAGGIO E, STORTI S F, GALAZZO I B, et al. Modulation of event-related desynchronization in robot-assisted hand performance: brain oscillatory changes in active, passive and imagined movements [J]. Journal of Neuroengineering and Rehabilitation, 2013, 10(1):1-10.
[13] MCFARLAND D J, MCCANE L M, DAVID S V, et al. Spatial filter selection for EEG-based communication [J]. Electroencephalography and Clinical Neurophysiology, 1997, 103(3): 386-394.
[14] SHEEHY N. Electroencephalography: basic principles, clinical applications, and related fields [J]. The Quarterly Review of Biology, 1983, 58(2):3108-3113.
[15] PFURTSCHELLER G, FH L D S. Event-related EEG/MEG synchronization and desynchronization: basic principles [J]. Clinical Neurophysiology, 1999, 110(11): 1842-1857.
[16] TOME D, MONTI F, BAROFFIO L, et al. Deep convolutional neural networks for pedestrian detection [J]. Signal Processing Image Communication, 2015,47(3): 482-489.
[17] TANG Z, LI C, SUN S. Single-trial EEG classification of motor imagery using deep convolutional neural networks [J]. Optik-International Journal for Light andElectron Optics, 2016, 130(1): 11-18.
[18] GENEVIEVE B O, MULLER K R. Neural networks: tricks of the trade [M]. Berlin: Springer, 2012:658-659.
[19] ROBINSON N, VINOD A P, ANG K K, et al. EEG based classification of fast and slow hand movements using wavelet-CSP algorithm [J]. IEEE Transactions on Biomedical Engineering, 2013, 60(8): 2123-2132.
[20] 李晓欧.基于独立分量分析和共同空间模式的脑电特征提取方法[J].生物医学工程学杂志,2010,27(6): 1370-1374. LI Xiao-ou. EEG feature extraction method based on independent component analysis and common space model [J]. Journal of Biomedical Engineering, 2010, 27(6): 1370-1374.
[21] HSU W Y, LIN C C, JU M S, et al. Wavelet-based fractal features with active segment selection: application to single-trial EEG data [J]. Journal of Neuroscience Methods, 2007, 163(1): 145-160.
[22] GHAEMI A, RASHEDI E, POURRAHIMI A M, et al. Automatic channel selection in EEG signals for classification of left or right hand movement in brain computer interfaces using improved binary gravitation search algorithm [J]. Biomedical Signal Processing and Control, 2017, 33(1): 109-118.
[23] SUBASI A. EEG signal classification using wavelet feature extraction and a mixture of expert model [J].Expert System with Application, 2007, 32(4):1084-1093.
[24] 杨新亮,罗志增.基于表面肌电信号的时频组合特征融合识别[J].华中科技大学学报:自然科学版,2011,39(2): 153-156. YANG Xin-liang, LUO Zhi-zeng. A recognition method of multi-channel SEMG based on the fusion of time-frequency combination characteristics [J]. Journal of Huazhong University of Science and Technology: Nature Science, 2011,39(2): 153-156.
[25] 王洁贞,韩兢,刘言训,等.Kappa统计量在一致性和重现性检验中的应用[J].山东大学学报:医学版,1996(3): 209-212. WANG Jie-zhen, HAN Jing, LIU Yan-xun, et al. Application of Kappa statistic in consistency and reproducibility test [J]. Journal of Shandong University: Health Sciences, 1996(3): 209-212.
[26] WANG B, JUN L, BAI J, et al. EEG recognition based on multiple types of information by using wavelet packet transform and neural networks [C]//International Conference of the IEEE Engineering in Medicine and Biology Society. Shanghai: IEEE, 2005:5377-5380.
[27] PFURTSCHELLER G, FH L D S. Event-related EEG/MEG synchronization and desynchronization: basic principles [J]. Clinical Neurophysiology, 1999, 110(11): 1842-1857.
[28] PFURTSCHELLER G, BRUNNER C, SCHLOGL A, et al. Mu rhythm (de) synchronization and EEG single-trial classification of different motor imagery tasks [J]. Neuroimage, 2006, 31(1): 153-159.
[29] 张志杰,刘强,黄希庭.时间知觉的神经机制-EEG时频分析的探索[J].西南大学学报:自然科学版,2007,29(10): 152-155. ZHANG Zhi-jie, LIU Qiang, HUANG Xi-ting. Mechanisms of time perception time-frequency analysis based on EEG data [J]. Journal of Southwest University: Natural Science Edition, 2007, 29(10): 152-155.
[30] PISTOHL T, SCHULZEBONHAGE A, AERTSEN A, et al. Decoding natural grasp types from human EcoG [J]. Neuroimage, 2011, 59(1): 248-260.
[31] MEIER J D, AFLALO T N, KASTNER S, et al. Complex organization of human primary motor cortex: a high-resolution fMRI study [J]. Journal of Neurophysiology, 2008, 100(4): 1800-1812.
[32] SCHIEBER M H, HIBBARD L S. How somatotopic is the motor cortex hand area? [J]. Science, 1993,261(5120): 489-492.
[33] LEMM S, BLANKERTZ B, CURIO G, et al. Spatio-spectral filters for improving the classification of single trial EEG [J]. IEEE Transactions on Bio-medical Engineering, 2005, 52(9): 1541-1548.
[34] SIVASANKARI K, THANUSHKODI K. An improved EEG signal classification using neural network with the consequence of ICA and STFT [J]. Journal of Electrical Engineering and Technology, 2014, 9(9): 1060-1071.

[1] 都明宇, 鲍官军, 杨庆华, 王志恒, 张立彬. 基于改进支持向量机的人手动作模式识别方法[J]. 浙江大学学报(工学版), 2018, 52(7): 1239-1246.
[2] 秦超, 梁喜凤, 路杰, 彭明, 金超杞. 七自由度番茄收获机械手的轨迹规划与仿真[J]. 浙江大学学报(工学版), 2018, 52(7): 1260-1266.
[3] 王禹, 韦巍. 采用频域Prony方法估计信号重叠双分量[J]. 浙江大学学报(工学版), 2018, 52(6): 1157-1166.
[4] 王扬威, 兰博文, 刘凯, 赵东标. 形状记忆合金丝驱动的柔性机械臂建模与实验[J]. 浙江大学学报(工学版), 2018, 52(4): 628-634.