Hi Doug,

Thank you for your reply. I computed the tkRAS coordinates using the formula TkRAS = Torig*[C R S 1]’. It returns me coordinates such as: [113, -119,88], [105, -103, 96]. I wonder whether such tkRAS coordinates are normal since

in label files obtained by saving ROI upon a certain threshold in tksurfer, the coordinates seem fairly different: non integers, and in different ranges e.g. [-36.806 -18.293 64.462], [-21.393 -61.263 8.910]) etc.

However, when I create a label file (via write_label.m) with -1 in the first column and with the TkRAS coordinates that I computed in the columns 2,3,4 , and then via tksurfer load this label and then save the selected label, I obtain a new label with R A S coordinates as [34.198 -48.599 29.835], [35.364 -45.689 31.179], i.e. similar to the one that I have in my ROI label files.

Could you please help me to figure out what my tkRAS values (e.g.[113, -119,88], [105, -103, 96) correspond to, versus the ones returned by tksurfer ([34.198 -48.599 29.835], [35.364 -45.689 31.179])?

I hope that it makes sense. Thank you so much!

Florence

From: Douglas Greve <greve@nmr.mgh.harvard.edu>
Date: July 24, 2017 at 4:06:25 PM EDT
To: <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] converting vertex numbers in .label files to R, A, S coordinates
Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>

Try using mri_label2label with the --paint option. I think the vertex numbers need to be -1 (not random numbers). You will also need to convert the RAS to "tkregisterRAS", one of our internal coordinate systems. See https://surfer.nmr.mgh.harvard.edu/fswiki/CoordinateSystems, you'll need to do the inverse of case #1.

On 7/19/17 12:28 PM, Li Guo wrote:
Dear Freesurfer experts,

My question concerns the .label files and the correspondence between the vertex numbers (in the first column) and the R,A,S coordinates.

I have the R,A,S coordinates of voxels that I want to display on the surface. For this purpose, I created a label file with the R,A,S coordinates (col 2,3,4) and random vertex ID in the col 1 (since I do not know what is the vertex ID for a given R,A,S coordinate). I displayed the content of the label file via tksurfer. However, tksurfer reads vertex ID, not R,A,S coordinates.

I wonder whether anyone knows about a function computing the vertex ID based on the R,A,S coordinates, or an alternative solution to represent the vertices on the surface based on the voxel coordinates--and the registration file I assume?

Thank you very much!

Best
Li Guo & Florence Campana