Classification of Schizophrenia and Depression by EEG with ANNs*

Ying-Jie Li; Fei-Yan Fan

doi:10.1109/IEMBS.2005.1617022

Classification of Schizophrenia and Depression by EEG with ANNs*

Conf Proc IEEE Eng Med Biol Soc. 2005:2005:2679-82. doi: 10.1109/IEMBS.2005.1617022.

Authors

Ying-Jie Li¹, Fei-Yan Fan

Affiliation

¹ Sch. of Commun. & Inf. Eng., Shanghai Univ.

PMID: 17282791
DOI: 10.1109/IEMBS.2005.1617022

Abstract

The clinical application shows that it is possible to differentiate between patients suffering from schizophrenia, depression and normal healthy persons on the basis of EEG rhythms. This paper describes the application of two artificial neural networks (ANN) approaches, BP ANN and self-organizing competitive ANN for the discrimination of three kinds of subjects (including 10 normal control, 10 schizophrenic patients and 10 depressive patients), with EEG rhythms used as feature vectors. In addition, the comparison between two ANNs is illustrated in this paper. The results show that ANN is an effective approach for discrimination of these three kinds of objects and BP ANN has better comprehensive performance than self-organizing competitive ANN technique in this study. Therefore, the ANN technique could be used as a new tool for computer-aided diagnosis for some psychosis.