Dear Freesurfers,
according to my research and ‘playing-around’ with Freesurfer tools on measuring distances between vertices or peak voxels, I’m still unsure about the
most adequate way of comparing distances between peak voxels corrected for folding patterns.
To my current knowledge, I figured out the following procedures:
Goal: comparing cortical distances between peak voxels between subjects of different clinical populations
1.)
Manual procedure using Freeview spline tool: using orig.mgz + white and pial surfaces + thresh_zstat images (FSL)
è
Using Measure tool (spline) to create a smoothed line within the cortical ribbon
è
Output: absolute cortical distances in [mm]
è
Problem: topologically remote but geometrically close by peak voxels (maybe avoidable by using the RAS to vertex transformation information of the inflated brain)
2.)
Automated procedure: using mris_pmake with start vertex stop vertex coordinates derived from the
population averaged sphere coordinate systems (fsaverage sphere) (functional data mapped into common sphere by either feat2surf or mris_preproc)
è
Output: only relative distance measures, but the comparability between peak voxel distances of different subjects is ensured (due to distance measures within a common spherical coordinate system)*.
è
Problem: no absolute measure of distances anymore (but this is actually no problem because I’m interested in relative differences in cortical reorganization)
- Do you have any suggestions which procedure is preferable??
- *Am I right about the comparability of distance measurements between subjects, when using mris_pmake within a common spherical coordinate space??
I appreciate any helpful comment
Sincerely yours,
Christopher
Christopher Milde, M.Sc. Biol.
Institute for Cognitive and Clinical Neuroscience
Central Institute of Mental Health Square J 5
68159 Mannheim, Germany
Phone: +49-621-1703-6313
E-mail: christopher.milde@zi-mannheim.de
Homepage:
http://www.zi-mannheim.de/
Office: Forschungs- und Verwaltungsgebäude, Room 230