Intracranial Current Signals Classification using Multivariate Autoregressive Modeling and Simulated Annealing Technique

C.E. Vasios, G.K. Matsopoulos, E.M. Ventouras, C.C. Papageorgiou, K.S. Nikita, V.P. Kontaxakis, G.N.

Keywords

ERPs, Intracranial Current Source,Multivariate Autoregression, Simulated Annealing,Classification, Neural Network

Abstract

Intracranial currents computed by the scalp-recorded ERPs, provide information on the non-observable electrical phenomena taking place in the brain, related to the cognitive mechanisms induced by the experimental task used in the ERP recording procedure. The use of current source waveforms, as input in classification systems, may provide robust classifiers due to the immediate relationship of the current sources to the brain electrical activity related to cognitive mechanisms. In the present work, a new method for the classification of intracranial current sources is proposed, combining the Multivariate Autoregressive model with the Simulated Annealing technique, in order to extract optimum features, in terms of the classification rate. The classification is implemented using a three-layer neural network (NN) trained with the back-propagation algorithm. The system was applied in the classification of normal controls and schizophrenic patients, providing classification rates of up to 100%. Furthermore, the clustering of intracranial source locations providing best classification performance may indicate relationships between the brain areas corresponding to these locations and pathological mechanisms.

Important Links:

Go Back