Novel neural network training methods (commonly referred to as deep learning) have emerged in recent years. Using a combination of unsupervised pre-training and subsequent fine-tuning, deep neural networks have become one of the most reliable classification methods. Since deep neural networks are especially powerful for high-dimensional and non-linear feature vectors, electroencephalography (EEG) and event-related potentials (ERPs) are one of the promising applications. Furthermore, to the authors' best knowledge, there are very few papers that study deep neural networks for EEG/ERP data. The aim of the experiments subsequently presented was to verify if deep learning-based models can also perform well for single trial P300 classification with possible application to P300-based brain-computer interfaces. The P300 data used were recorded in the EEG/ERP laboratory at the Department of Computer Science and Engineering, University of West Bohemia, and are publicly available. Stacked autoencoders (SAEs) were implemented and compared with some of the currently most reliable state-of-the-art methods, such as LDA and multi-layer perceptron (MLP). The parameters of stacked autoencoders were optimized empirically. The layers were inserted one by one and at the end, the last layer was replaced by a supervised softmax classifier. Subsequently, fine-tuning using backpropagation was performed. The architecture of the neural network was 209-130-100-50-20-2. The classifiers were trained on a dataset merged from four subjects and subsequently tested on different 11 subjects without further training. The trained SAE achieved 69.2% accuracy that was higher (p < 0.01) than the accuracy of MLP (64.9%) and LDA (65.9%). The recall of 58.8% was slightly higher when compared with MLP (56.2%) and LDA (58.4%). Therefore, SAEs could be preferable to other state-of-the-art classifiers for high-dimensional event-related potential feature vectors.

Neuroinformatics Research Group Department of Computer Science and Engineering Faculty of Applied Sciences University of West BohemiaPilsen Czechia

Zobrazit více v PubMed

Arnold L., Rebecchi S., Chevallier S., Paugam-Moisy H. (2011). An introduction to deep-learning, in Advances in Computational Intelligence and Machine Learning, ESANN'2011 (Bruges: ), 477–488.

Bengio Y., Lamblin P., Popovici D., Larochelle H. (2007). Greedy layer-wise training of deep networks, in Advances in Neural Information Processing Systems 19, eds Schölkopf B., Platt J., Hoffman T. (Vancouver: MIT Press; ), 153–160.

Blankertz B., Lemm S., Treder M. S., Haufe S., Müller K.-R. (2011). Single-trial analysis and classification of ERP components - a tutorial. Neuroimage 56, 814–825. 10.1016/j.neuroimage.2010.06.048 PubMed DOI

Blankertz B., Muller K., Curio G., Vaughan T., Schalk G., Wolpaw J., et al. . (2004). The BCI competition 2003: progress and perspectives in detection and discrimination of EEG single trials. IEEE Trans. Biomed. Eng. 51, 1044–1051. 10.1109/TBME.2004.826692 PubMed DOI

Cashero Z. (2012). Comparison of EEG Preprocessing Methods to Improve the Performance of the P300 Speller. Fort Collins: Proquest, Umi Dissertation Publishing.

Deng L., Yu D. (2013). Deep learning: methods and applications. Found. Trends Signal Process. 7, 197–387. 10.1561/2000000039 DOI

Dudacek K., Mautner P., Moucek R., Novotny J. (2011). Odd-ball protocol stimulator for neuroinformatics research, in Applied Electronics (AE), 2011 International Conference on (Pilsen: ), 1–4.

Farwell L. A., Donchin E. (1988). Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials. Electroencephalogr. Clin. Neurophysiol. 70, 510–523. 10.1016/0013-4694(88)90149-6 PubMed DOI

Fazel-Rezai R., Allison B. Z., Guger C., Sellers E. W., Kleih S. C., Kübler A. (2012). P300 brain computer interface: current challenges and emerging trends. Front. Neuroeng. 5:14. 10.3389/fneng.2012.00014 PubMed DOI PMC

Guger C., Daban S., Sellers E., Holzner C., Krausz G., Carabalona R., et al. . (2009). How many people are able to control a P300-based brain-computer interface (BCI)? Neurosci. Lett. 462, 94–98. 10.1016/j.neulet.2009.06.045 PubMed DOI

Haghighatpanah N., Amirfattahi R., Abootalebi V., Nazari B. (2013). A single channel-single trial p300 detection algorithm, in 2013 21st Iranian Conference on Electrical Engineering (ICEE) (Mashhad: ), 1–5.

Jansen B. H., Allam A., Kota P., Lachance K., Osho A., Sundaresan K. (2004). An exploratory study of factors affecting single trial p300 detection. IEEE Trans. Biomed. Eng. 51, 975–978. 10.1109/TBME.2004.826684 PubMed DOI

Ji S., Ye J. (2008). A unified framework for generalized linear discriminant analysis, in A unified framework for generalized linear discriminant analysis, (Anchorage, AK: ). PubMed

Krusienski D. J., Sellers E. W., Cabestaing F., Bayoudh S., McFarland D. J., Vaughan T. M., et al. . (2006). A comparison of classification techniques for the P300 Speller. J. Neural Eng. 3, 299–305. 10.1088/1741-2560/3/4/007 PubMed DOI

Lakey C. E., Berry D. R., Sellers E. W. (2011). Manipulating attention via mindfulness induction improves P300-based brain-computer interface performance. J. Neural Eng. 8:025019. 10.1088/1741-2560/8/2/025019 PubMed DOI PMC

Lotte F., Congedo M., Lécuyer A., Lamarche F., Arnaldi B. (2007). A review of classification algorithms for EEG-based brain–computer interfaces. J. Neural Eng. 4:24. 10.1088/1741-2560/4/2/R01 PubMed DOI

Luck S. (2005). An Introduction to the Event-Related Potential Technique. Cambridge, MA: MIT Press.

Manyakov N. V., Chumerin N., Combaz A., Van Hulle M. M. (2011). Comparison of classification methods for P300 brain-computer interface on disabled subjects. Comput. Intell. Neurosci. 2011:519868. 10.1155/2011/519868 PubMed DOI PMC

MATLAB (2015). MATLAB Version 8.6.0 (R2015b) - Neural Network Toolbox. Natick, MA: The MathWorks Inc.

Mirghasemi H., Fazel-Rezai R., Shamsollahi M. B. (2006). Analysis of P300 classifiers in brain computer interface speller. Conf. Proc. IEEE Eng. Med. Biol. Soc. 1, 6205–6208. 10.1109/IEMBS.2006.259521 PubMed DOI

Moucek R., Jezek P. (2009). EEG/ERP Portal. Available online at: http://eegdatabase.kiv.zcu.cz/

Ng A., Ngiam J., Foo C. Y., Mai Y., Suen C. (2010). UFLDL Tutorial. Available online at: http://ufldl.stanford.edu/wiki/index.php/UFLDL_Tutorial PubMed

Polich J. (2007). Updating P300: an integrative theory of P3a and P3b. Clin. Neurophysiol. 118, 2128–2148. 10.1016/j.clinph.2007.04.019 PubMed DOI PMC

Pound M. P., Burgess A. J., Wilson M. H., Atkinson J. A., Griffiths M., Jackson A. S., et al. (2016). Deep machine learning provides state-of-the-art performance in image-based plant phenotyping. bioRxiv. 10.1101/053033 Available online at: http://biorxiv.org/content/early/2016/05/12/053033 PubMed DOI PMC

Sellers E. W., Krusienski D. J., McFarland D. J., Vaughan T. M., Wolpaw J. R. (2006). A P300 event-related potential brain-computer interface (BCI): the effects of matrix size and inter stimulus interval on performance. Biol. Psychol. 73, 242–252. 10.1016/j.biopsycho.2006.04.007 PubMed DOI

Sobhani A. (2014). P300 Classification using Deep Belief Nets. Master's thesis, Colorado State University.

Thulasidas M., Guan C., Wu J. (2006). Robust classification of EEG signal for brain-computer interface. IEEE Trans. Neural Syst. Rehabil. Eng. 14, 24–29. 10.1109/TNSRE.2005.862695 PubMed DOI

Vareka L., Bruha P., Moucek R. (2014a). Event-related potential datasets based on a three-stimulus paradigm. Gigascience 3:35. 10.1186/2047-217X-3-35 PubMed DOI PMC

Vareka L., Bruha P., Moucek R. (2014b). Supporting Material for: “Event-Related Potential Datasets Based on Three-Stimulus Paradigm.” GigaScience Database. 10.5524/100111 PubMed DOI PMC

Vareka L., Mautner P. (2015). Using the Windowed means paradigm for single trial P300 detection, in 2015 38th International Conference on Telecommunications and Signal Processing (TSP) (Prague: ), 1–4.

Zamparo L., Zhang Z. (2015). Deep autoencoders for dimensionality reduction of high-content screening data. CoRR abs/1501.01348.