On 2/19/16 11:00 AM, Thomas Jacquemont wrote:
Dear FreeSurfer experts,
I would like to use the P. Hagmann atlas (Lausanne 2008 available here https://github.com/LTS5/cmp/tree/master/cmp/data/colortable_and_gcs) on my subjects. I ran Recon all on their T1 images and then these two commands to get the Hagmann parcellation for my subject:
mris_ca_label -dir ../subject_dir/003_S_2374 rh sphere.reg /aramis/home/jacquemont/DataAramis/Atlas/Hagmann/myatlas_250_rh.gcs ./test_sub_003_S_2374.rh.Hagmann500.annot
mri_label2vol --subject 003_S_2374 --hemi rh --annot ./test_sub_003_S_2374.rh.Hagmann500.annot --temp ../subject_dir/003_S_2374/mri/orig/001.mgz --o annoted_volume_Hagmann500_003_S_2374.mgz --identity
I have two questions. First, looking at the recon all log file I saw that the command is used this way:
#@# Cortical Parc lh mardi 19 janvier 2016, 23:32:45 (UTC+0100) mris_ca_label -l ../label/lh.cortex.label -aseg ../mri/aseg.mgz -seed 1234 NameSubject_M0 lh ../surf/lh.sphere.reg /path/to/freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0/average/lh.curvature.buckner40.filled.desikan_killiany.2010-03-25.gcs ../label/lh.aparc.annot
I would like to know what is the "seed" option? Do I need it?
There is some randomness built into mris_ca_label, and the seed is the seed for the random generator. We want the results to be dterministic (ie, gives the same answer when re-run in the same way) so we always seed with the same number. Probably you will want this too, so use the seed option.
Secondly, By surperimposed the resulting parcellation file (output of the mri_label2vol command) to the repective T1 image it seems that they are not in the same space. How could I have the pacellated file in the same space as the T1 image? Is the way I am using these two command good? What inputs should I change? The sphere file (for mris_ca_label), the --temp (for mri_label2vol), etc...
Use --regheader orig.mgz instead of --identity doug
Thank you for your help,
Yours faithfully,
Thomas
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer