Advertisement

Convolutional Neural Networks for Four-Class Motor Imagery Data Classification

  • Tomas Uktveris
  • Vacius Jusas
Chapter
Part of the Studies in Computational Intelligence book series (SCI, volume 737)

Abstract

In this paper the use of convolutional neural networks (CNN) is discussed in order to solve four class motor imagery classification problem. Analysis of viable CNN architectures and their influence on the obtained accuracy for the given task is argued. Furthermore, selection of optimal feature map image dimension, filter sizes and other CNN parameters used for network training is investigated. Methods for generating 2D feature maps from 1D feature vectors are presented for commonly used feature types. Initial results show that CNN can achieve high 68% classification accuracy for the four class motor imagery problem with less complex feature extraction techniques. It is shown that optimal accuracy highly depends on feature map dimensions, filter sizes, epoch count and other tunable factors, therefore various fine-tuning techniques must be employed. Experiments show that simple FFT energy map generation techniques are enough to reach the state-of-the-art classification accuracy for common CNN feature map sizes. This work also confirms that CNNs are able to learn a descriptive set of information needed for optimal electroencephalogram (EEG) signal classification.

Keywords

Convolutional neural network Motor imagery Feature map Image classification, FFT energy map 

References

  1. 1.
    Qin, L., He, B.: A wavelet-based time-frequency analysis approach for classification of motor imagery for brain-computer interface applications. J. Neural Eng. 2, 65–72 (2005)CrossRefGoogle Scholar
  2. 2.
    Xiao, D., Mu, Z.D., Hu, J. F.: Classification of motor imagery EEG signals based on energy entropy. In: International Symposium on Intelligent Ubiquitous Computing and Education, 61–64 (2009)Google Scholar
  3. 3.
    Zhou, B., Wu, X., Zhang, L., Lv, Z., Guo, X.: Robust spatial filters on three-class motor imagery EEG data using independent component analysis. J. Biosci. Med. 2, 43–49 (2014)Google Scholar
  4. 4.
    Bai, X., Wang, X., Zheng, S., Yu, M.: The offline feature extraction of four-class motor imagery EEG based on ICA and Wavelet-CSP. In: Control Conference (CCC), pp. 7189–7194 (2014)Google Scholar
  5. 5.
    Yang, H., Sakhavi, S., Ang, K.K., Guan, C.: On the use of convolutional neural networks and augmented CSP features for multi-class motor imagery of EEG signals classification. In: 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 2620–2623 (2015)Google Scholar
  6. 6.
    Naeem, M., Brunner, C., Pfurtscheller, G.: Dimensionality reduction and channel selection of motor imagery electroencephalographic data. Comput. Intell. Neurosci. (2009)Google Scholar
  7. 7.
    Wang, Y., Gao, S., Gao, X.: Common spatial pattern method for channel selection in motor imagery based brain-computer interface. In: IEEE Engineering in Medicine and Biology 27th Annual Conference, pp. 5392–5395 (2005)Google Scholar
  8. 8.
    Uktveris, T., Jusas, V.: Comparison of Feature Extraction Methods for EEG BCI Classification, Information and Software Technologies: 21st International Conference, pp. 81–92 (2015)Google Scholar
  9. 9.
    Brunner, C. et al.: BCI Competition 2008—Graz data set A (2008)Google Scholar
  10. 10.
    Vedaldi, A., Lenc, K.: MatConvNet—convolutional neural networks for MATLAB. In: Proceedings of the ACM International Conference on Multimedia (2015)Google Scholar
  11. 11.
    Tabar, Y. R., Halici, U.: A novel deep learning approach for classification of EEG motor imagery signals. J. Neural Eng. 14(1) (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  1. 1.Software Engineering DepartmentKaunas University of TechnologyKaunasLithuania

Personalised recommendations