ICANN '08 Proceedings of the 18th international conference on Artificial Neural Networks, Part II
Implicit emotional tagging of multimedia using EEG signals and brain computer interface
WSM '09 Proceedings of the first SIGMM workshop on Social media
Automatic recognition of alertness level from EEG by using neural network and wavelet coefficients
Expert Systems with Applications: An International Journal
Journal of Medical Systems
International Journal of Artificial Intelligence and Soft Computing
EEG signal classification using the event-related coherence and genetic algorithm
BICS'13 Proceedings of the 6th international conference on Advances in Brain Inspired Cognitive Systems
Tool support for software lookup table optimization
Scientific Programming
Hi-index | 0.00 |
EEG analysis has played a key role in the modeling of the brain'scortical dynamics, but relatively little effort has been devoted todeveloping EEG as a limited means of communication. If severalmental states can be reliably distinguished by recognizing patternsin EEG, then a paralyzed person could communicate to a device suchas a wheelchair by composing sequences of these mental states. EEGpattern recognition is a difficult problem and hinges on thesuccess of finding representations of the EEG signals in which thepatterns can be distinguished. In this article, we report on astudy comparing three EEG representations, the unprocessed signals,a reduced-dimensional representation using the Karhunen -Loève transform, and a frequency-based representation.Classification is performed with a two-layer neural networkimplemented on a CNAPS server (128 processor, SIMD architecture) byAdaptive Solutions, Inc. Execution time comparisons show over ahundred-fold speed up over a Sun Sparc 10. The best classificationaccuracy on untrained samples is 73% using the frequency-basedrepresentation.