The Use of Wavelet Transform as a Preprocessor for the Neural Network Classification of EEG Signals

C.D. Lopes, M.A. Zaro, and A.A. Susin (Brazil)


EEG, Artificial Neural Network, Wavelet Transform, Alcoholism.


In this study, Wavelet Transform (WT) is used to analyze EEG data as an input to a feed forward neural network for signal classification of individuals at high risk (HR) for alcoholism. We used two types of mother wavelets for the matheematical processing of EEG data: (a) Biorthogonal (Bior) and (b|) Daubechies (Db). The results show that the wavelet transform can be used to provide a better classification by artificial neural network (ANN). Both ANN, trained with wavelet coefficients of Bior and Db, provided good performances (70%) in the classification task.

Important Links:

Go Back