Hello,

I am new freesurfer user, I need to have the labels of parcellation on analyze format. I did recon all command on my subject, I see the very beautiful parcellation on freesurfer. After reading all the question/response on the freesurfer website, I write my own txt file describing the 35 labels: I just copy paste the data obtained from the variable colortable.struct_name and colortable.table: the output of read_annotation('rh.aparc.annot'). It seems working correctly, because the mri_annotation2label creates 35 files of all the labels. To check some of them I used tkmedit chuv44 brain.mgz and I loaded one label, it seems good. Is this correct?

But when I tried to export this labels to analyze format (readable with mricro for example), I use mri_label2vol:

Mri_label2vol –label ./subects/chuv44/mylab/rh.superiorfrontal.label –temp $SUBJECTS_DIR/chuv44/mri/orig/chuv44.img –o ./subjects/chuv44_rh_one.img

I had a warning saying that the mat file is missing, and the result orientation is not good.

I then open my analyze file with v=spm_vol('chuv44.img') and copy the v.mat as the mat file freesurfer asked for:

v.mat

ans =

    -2     0     0   129

     0     2     0  -129

     0     0     2   -69

     0     0     0     1

The command line Mri_label2vol –label ./subects/chuv44/mylab/rh.superiorfrontal.label –temp $SUBJECTS_DIR/chuv44/mri/orig/chuv44.img –o ./subjects/chuv44_rh_one.img   works then correctly and write in the end:

Analyze output matrix

    -2     0     0   129

     0     2     0  -129

     0     0     2   -69

     0     0     0     1

---------------------

INFO: set hdr.hist.orient to -1

But when I open the output file chuv44_rh_one.img (analyze) the format is not the same as my image: the axial slice is coronal and vice versa.

First question: Is the staff I did with my home-made txt labels_file is correct?

And do you know what I should do to obtain my label mask in correct orientation?

Thanks in advance

Leila Cammoun