Design and on-line evaluation of type-2 fuzzy logic system-based framework for handling uncertainties in BCI classification
2008 (English)In: 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2008, 4242-4245 p.Conference paper (Refereed)
The practical applicability of brain-computer interface (BCI) technology is limited due to its insufficient reliability and robustness. One of the major problems in this regard is the extensive variability and inconsistency of brain signal patterns, observed especially in electroencephalogram (EEG). This paper presents a fuzzy logic (FL) approach to the problem of handling of the resultant uncertainty effects. In particular, it outlines the design of a novel type-2 FL system (T2FLS) classifier within the framework of an EEG-based BCI, and examines its on-line applicability in the presence of shortand long-term nonstationarities of spectral EEG correlates of motor imagery (imagination of left vs. right hand movement). The developed system is shown to effectively cope with realtime constraints. In addition, a comparative post hoc analysis has revealed that the proposed T2FLS classifier outperforms conventional BCI methods, like LDA and SVM, in terms of the maximum classification accuracy (CA) rates by a relatively small, yet statistically significant, margin.
Place, publisher, year, edition, pages
2008. 4242-4245 p.
, IEEE Engineering in Medicine and Biology Society Conference Proceedings, ISSN 1557-170X
brain-computer interface, electroencephalography, motor imagery, fuzzy systems, classification
Other Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-90087DOI: 10.1109/IEMBS.2008.4650146ISI: 000262404502255ISBN: 978-142441815-2OAI: oai:DiVA.org:kth-90087DiVA: diva2:504136
30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08
QC 201203052012-02-182012-02-182012-03-05Bibliographically approved