Abstract #2975
Multiple Sclerosis Clinical Classification Based on DTI Fiber Analysis
Claudio Stamile 1 , Gabriel Kocevar 1 , Franoise Durand-Dubief 1,2 , Franois Cotton 1,3 , Carole Frindel 1 , Salem Hannoun 1 , and Dominique Sappey-Marinier 1,4
1
CREATIS (CNRS UMR5220 & INSERM U1044),
Universit Lyon 1, INSA-Lyon, Villeurbanne, France,
2
Service
de Neurologie A, Hpital Neurologique, Hospices Civils
de Lyon, Bron, France,
3
Service
de Radiologie, Centre Hospitalier Lyon-Sud, Hospices
Civils de Lyon, Pierre-Benite, France,
4
CERMEP
- Imagerie du Vivant, Universit de Lyon, Bron, France
In this work, we present a fully automated SVM method
for subject classification in three groups: healthy
control subjects, relapsing-remitting and primary
progressive Multiple Sclerosis (MS) patients based on
the diffusion information obtained from several WM fiber
bundles. The classification performance results suggest
that each WM fiber bundle contributes differently to the
classification of the MS clinical form. Moreover, we
show that the classification result depends on the
diffusion metrics used to study the fiber bundle. This
result could be useful to identify a specific diffusion
metric that better characterizes the WM fiber bundle.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
