Multi-atlas label propagation for accurate anatomical segmentation of rat brain images
Josiane YANKAM NJIWA 1 , Rolf Heckemann 2 , Nicolas Costes 3 , Sandrine Bouvard 3,4 , Caroline Bouillot 3 , Luc Zimmer 3,4 , and Alexander Hammers 5
Neurodis Foundation, Lyon, Rhone alpes,
of Gothenburg, Sweden, Gothenburg, Sweden,
du vivant, Lyon, France,
de Recherche en Neurosciences de Lyon, Lyon, France,
Foundation, lyon, Rhone, France
Accurate image registration that enables comparisons
within and between subjects and determination of how
regions are affected by pathological or physiological
processes remains challenging. Automated multi-atlas
based approaches have been proposed for the human brain.
They provide highly accurate structural segmentations by
propagating manual delineations from multiple atlases
and consolidating them in the space of a target image.
The aim of this work was to develop a corresponding
method for automatically defining ROIs on MR images of
rat brains. Visual assessments confirmed the success of
the segmentation method on normal as well as on
pathological MR rat brain images.
This abstract and the presentation materials are available to members only;
a login is required.