Sparse classification with MRI based markers for neuromuscular disease categorization

Katerina Gkirtzou 1, 2 Jean-François Deux 3 Guillaume Bassez 4 Aristeidis Sotiras 5 Alain Rahmouni 3 Thibault Varacca 3 Nikos Paragios 1, 6, 7, 8, 9, 10 Matthew Blaschko 1, 2
1 GALEN - Organ Modeling through Extraction, Representation and Understanding of Medical Image Content
Inria Saclay - Ile de France, Ecole Centrale Paris
2 CVN - Centre de vision numérique
3 Hôpital Henri Mondor
4 Centre de reference des maladies rares neuromusculaires, APHP, GH Henri Mondor, Creteil
5 SBIA - Section of Biomedical Image Analysis [Philadelphia]
6 MAS - Mathématiques Appliquées aux Systèmes - EA 4037
7 IMAGINE [Marne-la-Vallée]
LIGM - Laboratoire d'Informatique Gaspard-Monge, CSTB - Centre Scientifique et Technique du Bâtiment, ENPC - École des Ponts ParisTech
8 LIGM - Laboratoire d'Informatique Gaspard-Monge
9 UPEM - Université Paris-Est Marne-la-Vallée
10 ENPC - École des Ponts ParisTech
Abstract : In this paper, we present a novel method for disease classification between two patient populations based on features extracted from Magnetic Resonance Imaging (MRI) data. Anatomically meaningful features are extracted from structural data (T1- and T2-weighted MR images) and Diffusion Tensor Imaging (DTI) data, and used to train a new machine learning algorithm, the k-support SVM (ksup-SVM). The k-support regularized SVM has an inherent feature selection property, and thus it eliminates the requirement for a separate feature selection step. Our dataset consists of patients that suffer from facioscapulohumeral muscular dystrophy (FSH) and Myotonic muscular dystrophy type 1 (DM1) and our proposed method achieves a high performance. More specifically, it achieves a mean Area Under the Curve (AUC) of 0.7141 and mean accuracy 77% ± 0.013. Moreover, we provide a sparsity visualization of the features in order to indentify their discriminative value. The results suggest the potential of the combined use of MR markers to diagnose myopathies, and the general utility of the ksup-SVM. Source code is also available at https://gitorious.org/ksup-svm.
Keywords : myopathies Diffusion Tensor Imaging k-support regularized SVM
Document type :
Conference papers
Complete list of metadatas
https://hal.inria.fr/hal-00845126
Contributor : Katerina Gkirtzou <>
Submitted on : Thursday, July 18, 2013 - 2:00:24 PM
Last modification on : Friday, October 4, 2019 - 1:30:17 AM
Long-term archiving on : Wednesday, April 5, 2017 - 12:49:59 PM

Files

MLMI-8.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00845126, version 1

Collections

INRIA | EC-PARIS | ENPC | IMAGINE | PARISTECH | MAS | ECP-DR | CSTB | UPEC-UPEM | CVN | LIGM | ESIEE-PARIS | MICS | UPEC

Citation

Katerina Gkirtzou, Jean-François Deux, Guillaume Bassez, Aristeidis Sotiras, Alain Rahmouni, et al.. Sparse classification with MRI based markers for neuromuscular disease categorization. 4th International Workhop on Machine Learning in Medical Imaging, Sep 2013, Nagoya, Japan. ⟨hal-00845126⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

1000

Files downloads

602

Contact


archive-ouverte@inria.fr