Real-time single channel EEG motor imagery based Brain Computer Interface

This paper presents a motor imagery based Brain Computer Interface (BCI) that uses single channel EEG signal from the C3 or C4 electrode placed in the motor area of the head. Time frequency analysis using Short Time Fourier Transform (STFT) is used to compute spectrogram from the EEG data. The STFT is scaled to have gray level values on which Grey Co-occurrence Matrix (GLCM) is computed. Texture descriptors such as correlation, energy, contrast, homogeneity and dissimilarity are calculated from the GLCM matrices. The texture descriptors are used to train a logistic regression classifier which is then used to classify the left and right motor imagery signals. The single-channel motor imagery classification system is tested offline with different subjects. The average offline accuracy is 87.6%. An online BCI system is implemented in openViBE with the single channel classification scheme. The stimuli presentations and feedback are implemented in Python and integrated with the openViBe BCI system.

[1]  T.M. McGinnity,et al.  Comparative Analysis of Spectral Approaches to Feature Extraction for EEG-Based Motor Imagery Classification , 2008, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[2]  Wei-Yen Hsu,et al.  EEG-based motor imagery analysis using weighted wavelet transform features , 2009, Journal of Neuroscience Methods.

[3]  Ruimin Wang,et al.  Classification of Four-Class Motor Imagery Employing Single-Channel Electroencephalography , 2014, PloS one.

[4]  Fan Yang,et al.  A frequency-temporal-spatial method for motor-related electroencephalography pattern recognition by comprehensive feature optimization , 2012, Comput. Biol. Medicine.

[5]  Richard G. Baraniuk,et al.  Sparse Bilinear Logistic Regression , 2014, ArXiv.

[6]  Ahmed H. Tewfik,et al.  Adapting subject specific motor imagery EEG patterns in space-time-frequency for a brain computer interface , 2009, Biomed. Signal Process. Control..

[7]  Chivalai Temiyasathit,et al.  Increase performance of four-class classification for motor-imagery based brain-computer interface , 2014, 2014 International Conference on Computer, Information and Telecommunication Systems (CITS).

[8]  Yanhui Guo,et al.  Time–frequency texture descriptors of EEG signals for efficient detection of epileptic seizure , 2016, Brain Informatics.

[9]  Guillaume Gibert,et al.  OpenViBE: An Open-Source Software Platform to Design, Test, and Use BrainComputer Interfaces in Real and Virtual Environments , 2010, PRESENCE: Teleoperators and Virtual Environments.

[10]  Qi Xu,et al.  Fuzzy support vector machine for classification of EEG signals using wavelet-based features. , 2009, Medical engineering & physics.

[11]  Rui Ma,et al.  An analytic spatial filter and a hidden Markov model for enhanced information transfer rate in EEG-based brain computer interfaces , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[12]  Min-You Chen,et al.  A multi-class pattern recognition method for motor imagery EEG data , 2011, 2011 IEEE International Conference on Systems, Man, and Cybernetics.