RTU Research Information System
Latviešu English

Publikācija: A New Method of Medical Image Classification Using Neural Networks Committees

Publication Type Full-text conference paper published in other conference proceedings
Funding for basic activity Unknown
Defending: ,
Publication language English (en)
Title in original language A New Method of Medical Image Classification Using Neural Networks Committees
Field of research 2. Engineering and technology
Sub-field of research 2.2 Electrical engineering, Electronic engineering, Information and communication engineering
Authors Maksims Alekseičevs
Aleksandrs Glazs
Keywords Neural networks; Neural network ensemble; Plurality voting; Ensemble average; Weighted voting
Abstract The motivation for our work is to develop an efficient method of classification for medical images, which will be able to distinguish healthy patient from diseases one, using neural networks committees. In this work the possibility of using different committee methods is analyzed. It is shown, that for successfully solving the given task, any reviewed committee method may be used, because their results have the same reliability of classification.
Reference Alekseičevs, M., Glazs, A. A New Method of Medical Image Classification Using Neural Networks Committees. In: Biomedical Engineering: Proceedings of the 12th International Conference, Lithuania, Kaunas, 23-24 October, 2008. Kaunas: Technologija, 2008, pp.229-232. ISBN 978-9955-25-576-5.
ID 5059