Abstract #4424
Spinal cord gray and white matter segmentation using atlas deformation
Benjamin De Leener 1 , Augustin Roux 1 , Manuel Taso 2,3 , Virginie Callot 2,3 , and Julien Cohen-Adad 1,4
1
Institute of Biomedical Engineering,
Polytechnique Montreal, Montreal, Quebec, Canada,
2
Aix-Marseille
Universit, CNRS, CRMBM UMR 7339, Marseille, France,
3
AP-HM,
Ple dimagerie mdicale, Hopital de la Timone, CEMEREM,
Marseille, France,
4
Functional
Neuroimaging Unit, CRIUGM, Universit de Montral,
Montreal, Quebec, Canada
We propose a novel method for segmenting the spinal cord
gray and white matter using atlas-based deformation. The
main steps are: (i) registration of the MNI-Poly-AMU
template on a T2*-weighted image, (ii) deformation of
the gray matter probabilistic atlas from the template on
T2*-weighted images using SyN transformation (ANTs) and
(iii) applying the resulting warping field on the white
matter atlas. Results demonstrate accurate gray matter
segmentation when compared to manual segmentation. An
application of the smooth deformation-based segmentation
is the accurate parcellation of the atlas of white
matter tracts for quantifying multi-parametric MRI
metrics.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
