Classification of homologous human chromosomes using mutual information maximization

TitleClassification of homologous human chromosomes using mutual information maximization
Publication TypeConference Paper
Year of Publication2001
AuthorsMousavi, P., S. S. Fels, R. K. Ward, and P. M. Lansdorp
Conference NameImage Processing, 2001. Proceedings. 2001 International Conference on
Pagination845 -848 vol.2
Date Publishedoct.
Keywordscellular biophysics, feature extraction, homologous human chromosome classification, image classification, medical image processing, multi-feature analysis, mutual information maximization, neural nets, optimisation, unsupervised learning, unsupervised neural network architecture

Multi-feature analysis of human chromosome images is a major step towards classification of homologous chromosomes. An automatic quantitative classification method is proposed for homolog differentiation using multiple features. This method is based on mutual information maximization applied to an unsupervised neural network architecture. The neural network consists of separate modules which are trained to classify homologs using independent features. Mutual information is then maximized between the outputs of the modules forcing them to produce the same classification results, for a given chromosome. The proposed method was successfully applied to classify homologs of chromosome 16 with 100% accuracy


a place of mind, The University of British Columbia

Electrical and Computer Engineering
2332 Main Mall
Vancouver, BC Canada V6T 1Z4
Tel +1.604.822.2872
Fax +1.604.822.5949

Emergency Procedures | Accessibility | Contact UBC | © Copyright 2020 The University of British Columbia