On 11/23/15 10:18 PM, Guido Orlando Pascariello wrote:

Hi FS Expert

I have just looked for this subject a long the list but I haven't found anything about.

I have been working with FreeSurfer and I could get some segmentation from my T1 data. I used 'mri_segstats' to quantify volume of some subcortical region. Also I want to get intensity statistics from these regions, so I used --in parameter. In the 'recon-all' pipeline, you use 'norm.mgz' but.. . I wonder is not better to use nu.mgz in order to get intensity values more related to original data? If is not. Can you explain me what '-canorm' step does? Because I don't get it

The norm is probably the better volume because it has the most effective intensity normalization. --canorm performs the normalization by finding voxels that it thinks are extremely likely to get white matter. Assuming that all WM has the same intensity, it fits a smooth curve between the WM points using soap bubble smoothing. If then divides each voxel in the image by the value of the smooth curve at that voxel. doug

I'll be grateful if you can help Being all for now, I await a reply

Guido PASCARIELLO Facultad de Ingeniería - Universidad Nacional de Entre Ríos Argentina

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer