Home Computer Science Computational Diffusion MRI: MICCAI Workshop, Athens, Greece, October 2016
Pial and White Surfaces
Extraction of T1-weighted MR scans were processed using FreeSurfer  to obtain individual pial and white matter surfaces. Homologue vertices between pial/white surfaces and between individuals allow to easily navigate between subjects. Cortex parcellation was done using the aparc.a2009s Destrieux’s atlas  defining 152 regions shown on Fig. 2 detailed in the next section.
Individual Quantitative Maps of Neurite Density
As displayed on Fig. 2, the correspondence of vertices allowed to extract a distribution of fintra for each vertex by sampling the matched fi„tra map along the segment defined by each pair of corresponding (pial, white) vertices. Maps of local distributions of fintra were computed and statistics were inferred to build maps of average, standard deviation (stddev) and median offintra at the vertex level. In order to get fintra distributions at the scale of cortical areas, the vertex-wise distributions included in the considered area were merged together providing a microstructural signature of this area per subject. Figure 2 summarizes the cortical sampling pipeline developed to extract quantitative features on the cortex at the individual level.
Fig. 2 Cortical sampling pipeline for one subject
|< Prev||CONTENTS||Next >|