Cross-validation and neural network architecture selection for the classification of intracranial current sources

Vasios, CE; Matsopoulos, GK; Ventouras, EM; Nikita, KS; Uzunoglu, N

dc.contributor.author	Vasios, CE	en
dc.contributor.author	Matsopoulos, GK	en
dc.contributor.author	Ventouras, EM	en
dc.contributor.author	Nikita, KS	en
dc.contributor.author	Uzunoglu, N	en
dc.date.accessioned	2014-03-01T02:42:33Z
dc.date.available	2014-03-01T02:42:33Z
dc.date.issued	2004	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/31044
dc.subject	Back propagation	en
dc.subject	BET-ART inversion	en
dc.subject	Cross-validation	en
dc.subject	Multivariate Autoregressive	en
dc.subject	Neural Network	en
dc.subject	Simulated Annealing	en
dc.subject	Structure selection	en
dc.subject.other	Algorithms	en
dc.subject.other	Backpropagation	en
dc.subject.other	Bioelectric potentials	en
dc.subject.other	Decision support systems	en
dc.subject.other	Genetic algorithms	en
dc.subject.other	Noninvasive medical procedures	en
dc.subject.other	Positron emission tomography	en
dc.subject.other	Regression analysis	en
dc.subject.other	Simulated annealing	en
dc.subject.other	BET-ART inversion	en
dc.subject.other	Cross-validation	en
dc.subject.other	Multivariate autoregressive	en
dc.subject.other	Structure selection	en
dc.subject.other	Neural networks	en
dc.title	Cross-validation and neural network architecture selection for the classification of intracranial current sources	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/NEUREL.2004.1416561	en
heal.identifier.secondary	http://dx.doi.org/10.1109/NEUREL.2004.1416561	en
heal.publicationDate	2004	en
heal.abstract	In the present paper, a new methodological approach, for the classification of first episode schizophrenic patients (FES) against normal controls, is proposed. The first step of the methodology applied is the feature extraction, which is based on the combination of the Multivariate Autoregressive model with the Simulated Annealing technique, in order to extract optimum features, in terms of classification rate. The classification, as the second step of the methodology, is implemented by means of an Artificial Neural Network (ANN) trained with the back-propagation algorithm under ""leave-one-out cross-validation"". The ANN is a multi-layer perceptron, the architecture of which, is selected after a detailed search. The proposed methodology has been applied for the classification of FES patients and normal controls using as input signals the intracranial current sources obtained by the inversion of ERPs using an Algebraic Reconstruction Technique. Results by implementing the proposed methodology provide classification rates of up to 93%. © 2004 IEEE.	en
heal.journalName	2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL 2004	en
dc.identifier.doi	10.1109/NEUREL.2004.1416561	en
dc.identifier.spage	151	en
dc.identifier.epage	158	en