CFP last date
20 May 2024
Call for Paper
June Edition
IJCA solicits high quality original research papers for the upcoming June edition of the journal. The last date of research paper submission is 20 May 2024

Submit your paper
Know more
The week's pick
Responsive image
Enhancing Privacy Preservation: Multi-Attribute Protection with P-Sensitive K-Anonymity

Twinkle Patel Kiran Amin
Random Articles
Reseach Article

Optimized DNN Classification Framework based on Filter Bank Common Spatial Pattern (FBCSP) for Motor-imagery-based BCI

by Zayyanu Shuaibu, Li Qi
journal cover thumbnail

International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Volume 175 - Number 15
Year of Publication: 2020
Authors: Zayyanu Shuaibu, Li Qi
10.5120/ijca2020920646

Zayyanu Shuaibu, Li Qi . Optimized DNN Classification Framework based on Filter Bank Common Spatial Pattern (FBCSP) for Motor-imagery-based BCI. International Journal of Computer Applications. 175, 15 ( Aug 2020), 16-25. DOI=10.5120/ijca2020920646

@article{ 10.5120/ijca2020920646,
author = { Zayyanu Shuaibu, Li Qi },
title = { Optimized DNN Classification Framework based on Filter Bank Common Spatial Pattern (FBCSP) for Motor-imagery-based BCI },
journal = { International Journal of Computer Applications },
issue_date = { Aug 2020 },
volume = { 175 },
number = { 15 },
month = { Aug },
year = { 2020 },
issn = { 0975-8887 },
pages = { 16-25 },
numpages = {9},
url = { https://ijcaonline.org/archives/volume175/number15/31529-2020920646/ },
doi = { 10.5120/ijca2020920646 },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
}
%0 Journal Article
%1 2024-02-07T00:25:07.378931+05:30
%A Zayyanu Shuaibu
%A Li Qi
%T Optimized DNN Classification Framework based on Filter Bank Common Spatial Pattern (FBCSP) for Motor-imagery-based BCI
%J International Journal of Computer Applications
%@ 0975-8887
%V 175
%N 15
%P 16-25
%D 2020
%I Foundation of Computer Science (FCS), NY, USA
Abstract

The Brain-Computer Interface (BCI) is a kind of communication channel between the brain and the outside world that is not dependent on the nervous system. The recent improvement of deep neural network (DNN) in a classification task lead several researchers to apply DNN methods for BCI motor imagery (MI) classification. However, DNN based classification methods for MI classification either record low accuracy or take a long time to classify MI signal due to the high dimensional nature of the MI signal. This issue prevents the DNN based MI systems from being deployed in a real application. An ideal DNN classification framework for the MI classification was proposed in this study, which can be trained with better accuracy in a short period. To address the long DNN training time and the challenges of low classification accuracy, a new method was proposed that uses the FBCSP method to reduce the data dimension and increase class discrimination by extracting the best subject-specific features of the CSP from raw EEG data as input into the DNN. The DNN is designed with a few layers, and several DNN hyperparameters have been evaluated. A configuration that has performed better in terms of minimum training time and good classification accuracy has been selected. To investigate the accuracy of the proposed DNN methods called FBCSP-DNN; first, we use traditional approaches such as LDA, SVM, and KNN to create a baseline; second, we choose some DNN studies used in other studies that have been applied to the same dataset used in this study; third, we look at the use of transfer learning, a network pre-training methodology for one data collection of some subjects and then fine-tuning for another to improve some subjects with low accuracy; finally, the results obtained using the proposed were compared to traditional methods and other competing DNN methods used in other studies. The performance of the proposed FBSCP-DNN method is evaluated using BCI competition IV dataset 2a. The results show that the proposed method (81.43%) is superior to the traditional classification methods (SVM: 72.37%; KNN: 61.06%; LDA: 72.07%). We also compared the proposed DNN method and other DNN methods proposed by other studies using the same dataset. The proposed method outperforms other studies in terms of accuracy of the MI classification for within-subjects and cross-subjects, which improved by 6% and 9% respectively. The FBCSP-DNN classification framework proposed in this paper has the advantages of short training time and high classification accuracy, which ensures reliability for the practical application of BCI-motor imagery systems.

References
  1. W. J. R. et al., “{B}rain-computer interface technology: a review of the first international meeting,” IEEE Trans. Rehabil. Eng., vol. 8, no. 2, pp. 164–173, 2000.
  2. F. Lotte, “A Tutorial on EEG Signal Processing Techniques for Mental State Recognition in Brain-Computer Interfaces. Eduardo Reck Miranda; Julien Castet. Guide to Brain-Computer Music Interfacing,” 2014.
  3. H. Lu, H. Eng, C. Guan, and S. Member, “2010 Regularized common spatial pattern with aggregation for EEG classification in small-sample setting.pdf,” vol. 57, no. 12, pp. 1–10, 2010.
  4. G. Schalk, J. Mellinger, G. Schalk, and J. Mellinger, “Brain Sensors and Signals,” A Pract. Guide. toBrain-computerInterfacing with BCI2000, pp. 9–35, 2010.
  5. F. Lotte, E. Signal, and C. Techniques, “Study of Electroencephalographic Signal Processing and Classification Techniques towards the use of Brain-Computer Interfaces in Virtual Reality Applications To cite this version?: HAL Id?: tel-00356346 Thèse l ’ Institut National des Sciences Appliquées,” 2009.
  6. S. Zhang, S. Wang, D. Zheng, K. Zhu, and M. Dai, “A novel pattern with high-level commands for encoding motor imagery-based brain-computer interface,” Pattern Recognit. Lett., vol. 125, pp. 28–34, 2019.
  7. S. U. Amin, M. Alsulaiman, G. Muhammad, M. A. Mekhtiche, and M. Shamim Hossain, “Deep Learning for EEG motor imagery classification based on multi-layer CNN's feature fusion,” Futur. Gener. Comput. Syst., vol. 101, pp. 542–554, 2019.
  8. H. Kang and S. Choi, “Bayesian common spatial patterns for multi-subject EEG classification,” Neural Networks, vol. 57, pp. 39–50, 2014.
  9. N. Brodu, F. Lotte, and A. Lécuyer, “Comparative study of band-power extraction techniques for Motor Imagery classification,” IEEE SSCI 2011 - Symp. Ser. Comput. Intell. - CCMB 2011 2011 IEEE Symp. Comput. Intell. Cogn. Algorithms, Mind, Brain, pp. 95–100, 2011.
  10. H. Cho, M. Ahn, K. Kim, and S. Chan Jun, “Increasing session-to-session transfer in a brain-computer interface with on-site background noise acquisition,” J. Neural Eng., vol. 12, no. 6, p. 66009, 2015.
  11. S. M. Plis et al., “Deep learning for neuroimaging: A validation study,” Front. Neurosci., vol. 8, no. 8 JUL, pp. 1–11, 2014.
  12. M. Alamgir, M. Grosse-Wentrup, and Y. Altun, “Multitask learning for brain-computer interfaces,” J. Mach. Learn. Res., vol. 9, pp. 17–24, 2010.
  13. C. T. Vi and S. Subramanian, “Detecting Error-Related Negativity for interaction design,” Conf. Hum. Factors Comput. Syst. - Proc., pp. 493–502, 2012.
  14. I. Iturrate, L. Montesano, and J. Minguez, “Shared-control brain-computer interface for a two-dimensional reaching task using EEG error-related potentials,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS, pp. 5258–5262, 2013.
  15. R. Chavarriaga, I. Iturrate, Q. Wannebroucq, and J. D. R. Millan, “Decoding fast-paced error-related potentials in monitoring protocols,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS, vol. 2015-Novem, pp. 1111–1114, 2015.
  16. J. Oxides, I. Iturrate, J. Minguez, and L. Montesano, “Analysis and asynchronous detection of gradually unfolding errors during monitoring tasks,” J. Neural Eng., vol. 12, no. 5, p. 56001, 2015.
  17. S. Fazli, F. Popescu, M. Danóczy, B. Blankertz, K. R. Müller, and C. Grozea, “Subject-independent mental state classification in single trials,” Neural Networks, vol. 22, no. 9, pp. 1305–1312, 2009.
  18. J. Thomas, T. Maszczyk, N. Sinha, T. Kluge, and J. Dauwels, “Deep learning-based classification for brain-computer interfaces,” 2017 IEEE Int. Conf. Syst. Man, Cybern. SMC 2017, vol. 2017-Janua, pp. 234–239, 2017.
  19. Y. R. Tabar and U. Halici, “A novel deep learning approach for classification of EEG motor imagery signals,” J. Neural Eng., vol. 14, no. 1, p. 16003, 2017.
  20. H. Dose, J. S. Møller, H. K. Iversen, and S. Puthusserypady, “An end-to-end deep learning approach to MI-EEG signal classification for BCIs,” Expert Syst. Appl., vol. 114, pp. 532–542, 2018.
  21. Z. Tang, C. Li, and S. Sun, “Single-trial EEG classification of motor imagery using deep convolutional neural networks,” Optik (Stuttg)., vol. 130, pp. 11–18, 2017.
  22. S. Sakhavi, C. Guan, and S. Yan, “Learning Temporal Information for Brain-Computer Interface Using Convolutional Neural Networks,” IEEE Trans. Neural Networks Learn. Syst., vol. 29, no. 11, pp. 5619–5629, 2018.
  23. R. Alazrai, M. Abuhijleh, H. Alwanni, and M. I. Daoud, “A Deep Learning Framework for Decoding Motor Imagery Tasks of the Same Hand Using EEG Signals,” IEEE Access, vol. 7, pp. 109612–109627, 2019.
  24. A. Yuksel and T. Olmez, “A neural network-based optimal spatial filter design method for motor imagery classification,” PLoS ONE, vol. 10, no. 5. 2015.
  25. F. Lotte et al., “A review of classification algorithms for EEG-based brain-computer interfaces: A 10-year update,” J. Neural Eng., vol. 15, no. 3, 2018.
  26. I. B. C. Interface et al., “Weighted Transfer Learning for Improving Motor,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 27, no. 7, pp. 1–34, 2019.
  27. X. An, D. Kuang, X. Guo, Y. Zhao, and L. He, “A deep learning method for classification of EAn, X., Kuang, D., Guo, X., Zhao, Y., & He, L. (2014). A deep learning method for classification of EEG data based on motor imagery. International Conference on. Retrieved from http://link.springer.com/chapter,” Int. Conf., pp. 203–210, 2014.
  28. N. Lu, T. Li, X. Ren, and H. Miao, “A Deep Learning Scheme for Motor Imagery Classification based on Restricted Boltzmann Machines,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 6, pp. 566–576, 2017.
  29. S. Bach, A. Binder, G. Montavon, F. Klauschen, K. R. Müller, and W. Samek, “On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation,” PLoS One, vol. 10, no. 7, pp. 1–46, 2015.
  30. K. K. Ang, Z. Y. Chin, H. Zhang, and C. Guan, “Filter Bank Common Spatial Pattern (FBCSP) in brain-computer interface,” Proc. Int. Jt. Conf. Neural Networks, pp. 2390–2397, 2008.
  31. S. Kumar, A. Sharma, and T. Tsunoda, “An improved discriminative filter bank selection approach for motor imagery EEG signal classification using mutual information,” BMC Bioinformatics, vol. 18, no. Suppl 16, 2017.
  32. S. Kumar and A. Sharma, “A new parameter tuning approach for enhanced motor imagery EEG signal classification,” Med. Biol. Eng. Comput., vol. 56, no. 10, pp. 1861–1874, 2018.
  33. B. Blankertz, R. Tomioka, S. Lemm, M. Kawanabe, and K.-R. Muller, “Filters for Robust EEG,” IEEE Signal Process. Mag., vol. January, no. 2008, pp. 41–56, 2008.
  34. C. Brunner, M. Naeem, R. Leeb, B. Graimann, and G. Pfurtscheller, “Spatial filtering and selection of optimized components in four class motor imagery EEG data using independent components analysis,” Pattern Recognit. Lett., vol. 28, no. 8, pp. 957–964, 2007.
  35. M. Naeem, C. Brunner, R. Leeb, B. Graimann, and G. Pfurtscheller, “Separability of four-class motor imagery data using independent components analysis,” J. Neural Eng., vol. 3, no. 3, pp. 208–216, 2006.
  36. K. K. Ang, Z. Y. Chin, H. Zhang, and C. Guan, “Robust Filter Bank Common Spatial Pattern (RFBCSP) in motor-imagery-based brain-computer interface,” Proc. 31st Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. Eng. Futur. Biomed. EMBC 2009, no. January 2014, pp. 578–581, 2009.
  37. R. T. Schirrmeister et al., “Deep learning with convolutional neural networks for EEG decoding and visualization,” Hum. Brain Mapp., vol. 38, no. 11, pp. 5391–5420, 2017.
Index Terms

Computer Science
Information Sciences

Keywords

Brain-Computer Interface (BCI) Deep Neural Network (DNN) Motor Imagery (MI) Filter Bank Common Spatial Pattern (FBCSP) electroencephalography (EEG).

Powered by PhDFocusTM