Hello FreeSurfer experts,
I am using freesurfer v5.3 on a linux os. I have a few questions regarding individual aseg statistics (volume, intensity) for a subcortical region that is not automatically segmented by freesurfer. Using tkmedit I have manually traced my subcortical ROI and saved it as a label. By looking at the label for each subject (i.e., less <subjid>/label/ROI.label ) I can see how many voxels are included in the label, the coordinates of each voxel in the label and the signal intensity of each of these voxels.
1. Is there anyway to transform the information in my label to individual statistics in a format similar to what I can obtain with the automatic segmentations and asegstats2table? And what additional steps are necessary to accomplish this?
2. Does the volume that was loaded when the label was created have an impact on any of the volume statistics computed? For example, I generated my label on each subject with the orig volume loaded (i.e., tkmedit <subjid> orig.mgz ) - if I label the same voxels on the norm.mgz, will the intensity values generated for each subject be different? And should I have used a different volume to create my labels?
3. In the label, what coordinate system is used to identify the labeled voxels? Are these coordinates in native space?
Thanks, Eileen.
Sent from my iPad
On 3/23/14 6:46 AM, Eileen Moore wrote:
Hello FreeSurfer experts,
I am using freesurfer v5.3 on a linux os. I have a few questions regarding individual aseg statistics (volume, intensity) for a subcortical region that is not automatically segmented by freesurfer. Using tkmedit I have manually traced my subcortical ROI and saved it as a label. By looking at the label for each subject (i.e., less <subjid>/label/ROI.label ) I can see how many voxels are included in the label, the coordinates of each voxel in the label and the signal intensity of each of these voxels.
- Is there anyway to transform the information in my label to individual statistics in a format similar to what I can obtain with the automatic segmentations and asegstats2table? And what additional steps are necessary to accomplish this?
Yes, use mri_label2vol to convert the label into a volume (run it with --help to get command line help and examples), then use mri_segstats passing the output of label2vol with the --seg option and specifying --id 1 and specifying --in norm.mgz
- Does the volume that was loaded when the label was created have an impact on any of the volume statistics computed? For example, I generated my label on each subject with the orig volume loaded (i.e., tkmedit <subjid> orig.mgz ) - if I label the same voxels on the norm.mgz, will the intensity values generated for each subject be different? And should I have used a different volume to create my labels?
It will not make a difference (only if you pass orig.mgz instead of norm.mgz to mri_segstats above).
- In the label, what coordinate system is used to identify the labeled voxels? Are these coordinates in native space?
They are not in native space. They are in something we call "tkregister space". If you search our wiki for "cooridinate systems" you will find some information about it doug
Thanks, Eileen.
Sent from my iPad _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas Greve -
Eileen Moore