%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi@80/2006/07.07.18.53
%2 sid.inpe.br/sibgrapi@80/2006/07.07.18.53.28
%@doi 10.1109/SIBGRAPI.2006.19
%T Extracting Discriminative Information from Medical Images: A Multivariate Linear Approach
%D 2006
%A Thomaz, Carlos,
%A Aguiar, Nelson,
%A Oliveira, Sergio,
%A Duran, Fabio,
%A Busatto, Geraldo,
%A Gillies, Duncan,
%A Rueckert, Daniel,
%@affiliation Department of Electrical Engineering, Centro Universitario da FEI, São Paulo, Brazil
%@affiliation Department of Electrical Engineering, Centro Universitario da FEI, São Paulo, Brazil
%@affiliation Department of Electrical Engineering, Centro Universitario da FEI, São Paulo, Brazil
%@affiliation Departments of Psychiatry and Radiology, Faculty of Medicine, University of São Paulo, Brazil
%@affiliation Departments of Psychiatry and Radiology, Faculty of Medicine, University of São Paulo, Brazil
%@affiliation Department of Computing, Imperial College, London, UK
%@affiliation Department of Computing, Imperial College, London, UK
%E Oliveira Neto, Manuel Menezes de,
%E Carceroni, Rodrigo Lima,
%B Brazilian Symposium on Computer Graphics and Image Processing, 19 (SIBGRAPI)
%C Manaus, AM, Brazil
%8 8-11 Oct. 2006
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K Statistical pattern recognition, Medical image computing.
%X Statistical discrimination methods are suitable not only for classification but also for characterisation of differences between a reference group of patterns and the population under investigation. In the last years, statistical methods have been proposed to classify and analyse morphological and anatomical structures of medical images. Most of these techniques work in high-dimensional spaces of particular features such as shapes or statistical parametric maps and have overcome the difficulty of dealing with the inherent high dimensionality of medical images by analysing segmented structures individually or performing hypothesis tests on each feature separately. In this paper, we present a general multivariate linear framework to identify and analyse the most discriminating hyper-plane separating two populations. The goal is to analyse all the intensity features simultaneously rather than segmented versions of the data separately or feature-by-feature. The conceptual and mathematical simplicity of the approach, which pivotal step is spatial normalisation, involves the same operations irrespective of the complexity of the experiment or nature of the data, giving multivariate results that are easy to interpret. To demonstrate its performance we present experimental results on artificially generated data set and real medical data.
%@language en
%3 thomaz-multivariate.pdf