Art neural networks for brain pathology diagnosis

Abstract

The present paper is devoted to the questions of the electroencephalograms (EEG) classification. The main objective of the work is ART Neural Network method as applied to the EEG clustering. The first section reviews the problem of epilepsy diagnostics. Then the algorithm based on the ART model is described. Adaptive Resonance Theory ( ART1) Neural Networks for fast, stable learning and prediction have been applied in a variety of areas. Applications include automatic target recognition, medical diagnosis. The paper describes ART1 model for recognition EEG patterns while diagnosing brain diseases.