Divergence-based framework for common spatial patterns algorithms

被引:131
作者
机构
[1] Berlin Institute of Technology, TU Berlin
[2] Advanced Telecommunication Research Institute International
[3] Department of Brain and Cognitive Engineering, Korea University
来源
| 1600年 / Institute of Electrical and Electronics Engineers Inc., United States卷 / 07期
基金
日本学术振兴会; 新加坡国家研究基金会;
关键词
Brain-computer interfaces; information geometry; spatial filters;
D O I
10.1109/RBME.2013.2290621
中图分类号
学科分类号
摘要
Controlling a device with a brain-computer interface requires extraction of relevant and robust features from high-dimensional electroencephalographic recordings. Spatial filtering is a crucial step in this feature extraction process. This paper reviews algorithms for spatial filter computation and introduces a general framework for this task based on divergence maximization. We show that the popular common spatial patterns (CSP) algorithm can be formulated as a divergence maximization problem and computed within our framework. Our approach easily permits enforcing different invariances and utilizing information from other subjects; thus, it unifies many of the recently proposed CSP variants in a principled manner. Furthermore, it allows to design novel spatial filtering algorithms by incorporating regularization schemes into the optimization process or applying other divergences. We evaluate the proposed approach using three regularization schemes, investigate the advantages of beta divergence, and show that subject-independent feature spaces can be extracted by jointly optimizing the divergence problems of multiple users. We discuss the relations to several CSP variants and investigate the advantages and limitations of our approach with simulations. Finally, we provide experimental results on a dataset containing recordings from 80 subjects and interpret the obtained patterns from a neurophysiological perspective. © 2008-2011 IEEE.
引用
收藏
页码:50 / 72
页数:22
相关论文
共 99 条
[21]  
Berger H., Ü ber das Elektrenkephalogramm des Menschen II, J. FürPsychologie und Neurologie, 40, pp. 160-179, (1930)
[22]  
Pfurtscheller G., Lopes Da Silva F.H., Event-related EEG/MEG synchronization and desynchronization: Basic principles, Clinical Neurophysiology, 110, 11, pp. 1842-1857, (1999)
[23]  
Niedermeyer E., Da Silva F.L., Electroencephalography: Basic PrinciplesClinical Applications and Related Fields, (2004)
[24]  
Blankertz B., Lemm S., Treder M.S., Et al., Singletrialanalysis and classification of ERP components-A tutorial, NeuroImage, 56, 2, pp. 814-825, (2011)
[25]  
Michel C.M., Murray M.M., Lantz G., Gonzalez S., Spinelli L., Grave De Peralta R., EEG source imaging, Clinical Neurophysiology, 115, 10, pp. 2195-2222, (2004)
[26]  
Lu H., Plataniotis K., Venetsanopoulos A., Regularized commonspatial patterns with generic learning for EEG signal classification, Proc. IEEE Int. Conf. Eng. Med. Biol. Soc., pp. 6599-6602, (2009)
[27]  
Kawanabe M., Vidaurre C., Improving BCI performance by modifiedcommon spatial patterns with robustly averaged covariance matrices, Proc. Int. Work-Conf. Artif. Neural Netw., pp. 279-282, (2009)
[28]  
Kawanabe M., Samek W., Muller K.-R., Vidaurre C., Robust commonspatial filters with a maxmin approach, Neural Comput., 26, 2, pp. 1-28, (2014)
[29]  
Krauledat M., Analysis of nonstationarities in EEG signals for improvingbrain-computer interface performance, Dept. Softw.Eng. Theoretical Comput. Sci., (2008)
[30]  
Bamdadian A., Guan C., Ang K.K., Xu J., Online semi-supervisedlearning with KL distance weighting for motor imagery-based BCI, Proc. IEEE Int. Conf. Eng. Med. Biol. Soc., pp. 2732-2735, (2012)