Brain-Computer Interface with Corrupted EEG Data: a Tensor Completion Approach
One of the current issues in brain-computer interface (BCI) is how to deal with noisy electroencephalography (EEG) measurements organized as multidimensional datasets (tensors). On the other hand, recently, significant advances have been made in multidimensional signal completion algorithms that exploit tensor decomposition models to capture the intricate relationship among entries in a multidimensional signal. We propose to use tensor completion applied to EEG data for improving the classification performance in a motor imagery BCI system with corrupted measurements. Noisy measurements (electrode misconnections, subject movements, etc.) are considered as unknowns (missing samples) that are inferred from a tensor decomposition model (tensor completion). We evaluate the performance of four recently proposed tensor completion algorithms, CP-WOPT (Acar et al. Chemom Intell Lab Syst. 106:41-56, 2011), 3DPB-TC (Caiafa et al. 2013), BCPF (Zhao et al. IEEE Trans Pattern Anal Mach Intell. 37(9):1751-1763, 2015), and HaLRT (Liu et al. IEEE Trans Pattern Anal Mach Intell. 35(1):208-220, 2013), plus a simple interpolation strategy, first with random missing entries and then with missing samples constrained to have a specific structure (random missing channels), which is a more realistic assumption in BCI applications. We measured the ability of these algorithms to reconstruct the tensor from observed data. Then, we tested the classification accuracy of imagined movement in a BCI experiment with missing samples. We show that for random missing entries, all tensor completion algorithms can recover missing samples increasing the classification performance compared to a simple interpolation approach. For the random missing channels case, we show that tensor completion algorithms help to reconstruct missing channels, significantly improving the accuracy in the classification of motor imagery (MI), however, not at the same level as clean data. Summarizing, compared to the interpolation case, all tensor completion algorithms succeed to increase the classification performance by 7–9% (LDA–SVD) for random missing entries and 15–8% (LDA–SVD) for random missing channels. Tensor completion algorithms are useful in real BCI applications. The proposed strategy could allow using motor imagery BCI systems even when EEG data is highly affected by missing channels and/or samples, avoiding the need of new acquisitions in the calibration stage.
KeywordsBrain-computer interface EEG Tensor completion Tensor decomposition Missing samples
JSC was supported by the Spanish Ministry of Science and Innovation (Grant No TEC2016-77791-C4-2-R) and the University of Vic – Central University of Catalonia (Grant No R0947). CFC was supported by NSF IIS-1636893, NSF BCS-1734853, NIH NIMH ULTTR001108, and partially supported by the Indiana University Areas of Emergent Research initiative “Learning: Brains, Machines, Children”. QZ was supported by JSPS KAKENHI Grant No. 17K00326, NSFC China Grant No. 61773129 and JST CREST Grant No. JPMJCR1784. AC was partially supported by the MES RF Grant No14.756.31.0001 and the Polish National Science Center Grant No 2016/20/W/N24/00354.
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflict of interest.
Informed consent was obtained from all participants.
Human and Animal Rights
All experiments were performed in accordance with the tenets of the Declaration of Helsinki.
- 4.Millán JD. Brain–computer interfaces. In Handbook of brain theory and neural networks. Cambridge: The MIT Press; 2002.Google Scholar
- 5.Huang D, Qian K, Fei D, Jia W, Chen X, Bai O. Electroencephalography (EEG)-based brain-computer interface (BCI): a 2-D virtual wheelchair control based on event-related desynchronization/synchronization and state control. IEEE Trans Neural Syst Rehabil Eng. 2012;20(3):379–88.PubMedCrossRefGoogle Scholar
- 10.Barmdad M, Zarshenas H, Auais M. Application of BCI systems in neurorehabilitation: a scoping review. Disabil Rehabil Assist Technol. 2015;5:1–10.Google Scholar
- 12.Millán J, Rupp R, Müller-Putz G, Murray-Smith R, Giugliemma C, Tangermann M, et al. Combining braincomputer interfaces and assistive technologies: state-of-the-art and challenges. Front Neurosci. 2010;7:161.Google Scholar
- 14.Lotte F, Congedo M, Lécuyer A, Lamarche F, Arnaldi B. A review of classification algorithms for EEG-based brain–computer interfaces. J Neural Eng. 2007;4(2).Google Scholar
- 18.Yang B, Zhang T, Zhang Y, Liu W, Wang J, Duan K. Removal of electrooculogram artifacts from electroencephalogram using canonical correlation analysis with ensemble empirical mode decomposition. Cogn Comput. 2017;1–8.Google Scholar
- 23.Sorber L, Van Barel M, De Lathauwer L. Tensorlab a MATLAB toolbox for tensor computations. 2014. [En línea]. Available: http://www.tensorlab.net/. [Last Access: 15 July 2015].
- 28.Fadili MJ, Starck J-L, Murtagh F. Inpainting and zooming using sparse representations. J Comput Secur. 2008;2:64–79.Google Scholar
- 30.Xu Y, Yin W. A block coordinate descent method for regularized multiconvex optimization with applications to nonnegative tensor factorization and completion, zi. 2013;6(3):1758–1789.Google Scholar
- 31.Kasai H, Mishra B. Low-rank tensor completion: a Riemannian manifold preconditioning approach, de 33 rd International Conference on Machine Learning, New York. 2016.Google Scholar
- 34.Jain P, Oh S. Provable tensor factorization with missing data, de Advances in Neural Information Processing Systems, Montreal. 2014.Google Scholar
- 35.Yokota T, Hontani H. Simultaneous tensor completion and denoising by noise inequality constrained convex optimization. arXiv. 2018;1801:03299.Google Scholar
- 38.Gia C, Zhong G, Fu Y. Low-rank tensor learning with discriminant analysis for action classification and image recovery, de Twenty-Eighth AAAI Conference on Artificial Intelligence, Quebec. 2014.Google Scholar
- 40.Elad M. Sparse and redundant representations: from theory to applications in signal and image processing, Springer. 2010.Google Scholar
- 43.Zhang Y, Zhao Q, Zhou G, Jin J, Wang X, Cichocki A. Removal of EEG artifacts for BCI applications using fully Bayesian tensor completion, de The 41st IEEE International Conference on Acoustic, Speech and Signal Processing - ICASSP 2016, Shanghai. 2016.Google Scholar
- 44.Krzanowski WJ. Principles of multivariate analysis: a user’s perspective. New-York: Oxford University Press; 1988.Google Scholar
- 46.Lotte F. A tutorial on EEG signal-processing techniques for mentalstate recognition in brain–computer interfaces, de Guide to Brain- Computer Music Interfacing. London: Springer; 2014. p. 133–61.Google Scholar
- 48.Vialatte FB, Solé-Casals J, Maurice M, Latchoumane C, Hudson N, Wimalaratna S, Jeong J, Cichocki A. Improving the quality of EEG data in patients with Alzheimer’s disease using ICA, de Advances in Neuro-Information Processing. ICONIP 2008. Lecture Notes in Computer, vol. 5507, Berlin, Springer, pp. 979–986.Google Scholar