Forecasting the MagnetoEncephaloGram (MEG) of Epileptic Patients Using Genetically Optimized Neural Networks

In this work MagnetoEncephaloGram (MEG) recordings of epileptic patients were analyzed using a hybrid neural networks training algorithm. This algorithm combines genetic algorithms and a training method based on the localized Extended Kalman Filter (EKF), in order to evolve the structure and train Multi-Layered Perceptrons (MLP) networks. Our goal is to examine the predictability of the MEG signal on a short and long predicting horizon. Numerous experiments were conducted giving highly successful results.

• Type: Conference
• Year: 1999
• Where: GECCO
• Authors: Adam V. Adamopoulos, Efstratios F. Georgopoulos, Spiridon D. Likothanassis, Photios A. Anninos
• Keywords: Extended Kalman Filter, GECCO 1999, Networks Training Algorithm, Optimization, Train Multi-layered Perceptrons