Publications

sort by:  author date

Stacked Autoencoders for the P300 Component Detection

VAŘEKA, L., MAUTNER, P. Stacked Autoencoders for the P300 Component Detection. Frontiers in Neuroscience, 2017, roč. 11, č. 302, s. 1-9. ISSN: 1662-453X
Abstract PDF BibTeX

Using a combination of unsupervised pre-training and subsequent fine-tuning, deep neural networks have become one of the most reliable classification methods. 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 were implemented and compared with some of the currently most reliable state-of-the-art methods, such as LDA and multi-layer perceptron. The parameters of stacked autoencoders were optimized empirically. 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%).

Text, Speech and Dialogue, 20 th International Conference, TSD 2017

Ekštein, K., Konopík, M., Matoušek, V., Mouček, R. Text, Speech and Dialogue, 20 th International Conference, TSD 2017. Praha, 27.08.2017 - 31.08.2017.
Abstract PDF BibTeX

The TSD 2017 conference was held on August 27 - 31, 2017 in Prague, Czech Republic. It was focused primarely on the following topics language corpora and their utilizaion, speech recognition, natural language processing, speech synthesis, automatic dialogtue systems and related.

Application of Stacked Autoencoders to P300 Experimental Data

VAŘEKA, L., PROKOP, T., MOUČEK, R., MAUTNER, P., ŠTĚBETÁK, J. Application of Stacked Autoencoders to P300 Experimental Data. In Artificial Intelligence and Soft Computing. Cham: Springer, 2017. s. 187-198. ISBN: 978-3-319-59062-2 , ISSN: 0302-9743
Abstract PDF BibTeX

Deep learning has emerged as a new branch of machine learning in recent years. Some of the related algorithms have been reported to beat state-of-the-art approaches in many applications. The main aim of this paper is to verify one of the deep learning algorithms, specifically a stacked autoencoder, to detect the P300 component. This component, as a specific brain response, is widely used in the systems based on brain-computer interface. A simple brain-computer interface experiment more than 200 school-age participants was performed to obtain large datasets containing the P300 component. After feature extraction the collected data were split into the training and testing sets. State-of-the art BCI classifiers (such as LDA, SVM, or Bayesian LDA) were applied to the data and then compared with the results of stacked autoencoders.

Applying an Archetype-Based Approach to Electroencephalography/Event-Related Potential Experiments in the EEGBase Resource

PAPEŽ, V., MOUČEK, R. Applying an Archetype-Based Approach to Electroencephalography/Event-Related Potential Experiments in the EEGBase Resource. Frontiers in Neuroinformatics, 2017, roč. 11, č. 24, s. 1-13. ISSN: 1662-5196
Abstract PDF BibTeX

The purpose of this study is to investigate the feasibility of applying openEHR (an archetype-based approach for electronic health records representation) to modeling data stored in EEGBase, a portal for experimental electroencephalography/eventrelated potential (EEG/ERP) data management. The study evaluates re-usage of existing openEHR archetypes and proposes a set of new archetypes together with the openEHR templates covering the domain. The main goals of the study are to (i) link existing EEGBase data/metadata and openEHR archetype structures and (ii) propose a new openEHR archetype set describing the EEG/ERP domain since this set of archetypes currently does not exist in public repositories

Event-related potential data from a guess the number brain-computer interface experiment on school children

MOUČEK, R., VAŘEKA, L., PROKOP, T., ŠTĚBETÁK, J., BRŮHA, P. Event-related potential data from a guess the number brain-computer interface experiment on school children. Scientific Data, 2017, roč. 4, č. March 2017, s. 1-11. ISSN: 2052-4463
Abstract PDF BibTeX

Guess the number is a simple P300-based brain-computer interface experiment. Its aim is to ask the measured participant to pick a number between 1 and 9. Then, he or she is exposed to corresponding visual stimuli and experimenters try to guess the number thought while they are observing event-related potential waveforms on-line. 250 school-age children participated in the experiments that were carried out in elementary and secondary schools in the Czech Republic. Electroencephalographic data from three EEG channels (Fz, Cz, Pz) and stimuli markers were stored. Additional metadata about the participants were collected (gender, age, laterality, the number thought by the participant, the guess of the experimenters, and various interesting additional information). Consequently, we offer the largest publicly available odd-ball paradigm collection of datasets to neuroscientific and brain-computer interface community.

Our partners