A decision support system of evoked potentials for the classification of patients with first-episode schizophrenia

Vasios, C; Papageorgiou, C; Matsopoulos, GK; Nikita, KS; Uzunoglu, N

dc.contributor.author	Vasios, C	en
dc.contributor.author	Papageorgiou, C	en
dc.contributor.author	Matsopoulos, GK	en
dc.contributor.author	Nikita, KS	en
dc.contributor.author	Uzunoglu, N	en
dc.date.accessioned	2014-03-01T01:51:49Z
dc.date.available	2014-03-01T01:51:49Z
dc.date.issued	2002	en
dc.identifier.issn	14331055	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/26460
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-0141763722&partnerID=40&md5=4373f04210ef6039577ce4ee0aa2523b	en
dc.subject	Autoregression model	en
dc.subject	Classification	en
dc.subject	Event-related potentials	en
dc.subject	Feature extraction	en
dc.subject	First-episode schizophrenia	en
dc.subject	Neural network	en
dc.subject.other	adult	en
dc.subject.other	age	en
dc.subject.other	article	en
dc.subject.other	artificial neural network	en
dc.subject.other	clinical article	en
dc.subject.other	computer analysis	en
dc.subject.other	controlled study	en
dc.subject.other	decision support system	en
dc.subject.other	event related potential	en
dc.subject.other	evoked response	en
dc.subject.other	female	en
dc.subject.other	gender	en
dc.subject.other	human	en
dc.subject.other	information processing	en
dc.subject.other	male	en
dc.subject.other	patient coding	en
dc.subject.other	regression analysis	en
dc.subject.other	schizophrenia	en
dc.subject.other	socioeconomics	en
dc.subject.other	working memory	en
dc.title	A decision support system of evoked potentials for the classification of patients with first-episode schizophrenia	en
heal.type	journalArticle	en
heal.publicationDate	2002	en
heal.abstract	Background: Recently it has been shown that the second-pass parsing process of information processing, as indexed by the P600 component of event-related potentials (ERPs), elicited during a working memory (WM) test, is impaired in first episode schizophrenic (FES) patients. Objective: The purpose of this study is to develop a decision support system - based on artificial neural networks (ANN) technology - for the classification of patients experiencing FES compared to healthy controls, utilizing the P600. Method: We examined 14 FES patients and 23 healthy controls, matched for age, sex and educational level. The proposed system comprises two levels: the feature extraction level and the classification level. The former is based on the implementation of an autoregression model to estimate the corresponding coefficients, which form the input vector for the later level. The classification level consists of a multi-layer neural network. Results: The performance of the system in terms of classification rate has been tested for a total of 15 abductions of each subject and for a specific order of the autoregression model according to the modified Schwarz criterion. The best classification rate, up to 100% has been achieved for the (C4-T6)/2 abduction compared to the other abductions and for all the subjects. Furthermore, the performance of the classifier for this abduction is consistent against the other adductions and for all the specific orders of the autoregression model implemented. Conclusions: The findings indicate that activities related to the P600 component during a WM task and explored by the proposed system may be involved in FES. Additionally, the findings also indicate that this approach may significantly facilitate the computer-aided analysis of ERPs.	en
heal.journalName	German Journal of Psychiatry	en
dc.identifier.volume	5	en
dc.identifier.issue	3	en
dc.identifier.spage	78	en
dc.identifier.epage	84	en