Neuro-wavelet classifiers for EEG signals based on rough set methods

Since EEG is one of the most important sources of information in therapy of epilepsy, several researchers tried to address the issue of decision suppo rt for such a data. In our work we try to establish a tool for noise-resist ant classification of EEG signals. The data we deal with is connected to di ssemination of different kinds of epilepsy. By identifying features in the signal we want to provide an automatic system that will support a physician in the diagnosing process. By applying the wavelets, frequential analysis, rough sets and dynamic scaling in connection with simple neural network we obtained novel and reliable classifier architecture. Experiments prove tha t the proposed method provides extended robustness and generalisation abili ties as well as a possibility to directly interpret the results obtained. ( C) 2001 Elsevier Science B.V. All rights reserved.