Brain Tumor Vascular Network Segmentation from Micro-Tomography

Xavier Descombes (1) , Franck Plouraboue (2) , Abdelhakim El Boustani (2, 3) , Caroline Fonta (4) , Géraldine Leduc (5) , Raphael Serduc (5) , Timm Weitkamp (5)

1 CRISAM - Inria Sophia Antipolis - Méditerranée
2 IMFT - Institut de mécanique des fluides de Toulouse
3 ENSAT - Ecole Nationale des Sciences Appliquées [Tanger]
4 CERCO - Centre de recherche cerveau et cognition
5 ESRF - European Synchrotron Radiation Facility

Micro-tomography produces high resolution images of bio- logical structures such as vascular networks. In this paper, we present a new approach for segmenting vascular network into pathological and normal regions from considering their micro-vessel 3D structure only. We define and use a condi- tional random field for segmenting the output of a watershed algorithm. The tumoral and normal classes are thus character- ized by their respective distribution of watershed region size interpreted as local vascular territories.

Keywords: Brain Tumor; Vascular Network; Segmentation; Micro-Tomography

Domains: Biotechnology; Cancer; Fluids mechanics [physics.class-ph]

Xavier Descombes, Franck Plouraboue, Abdelhakim El Boustani, Caroline Fonta, Géraldine Leduc, et al.. Brain Tumor Vascular Network Segmentation from Micro-Tomography. 2011 IEEE International Symposium on Biomedical Imaging, Mar 2011, Chicago, United States. pp.1113-1116. ⟨hal-04041532⟩