Hi Linnea – Before the stats are computed, the probability distribution of the paths is thresholded to 20% of its peak value, so voxels with very low probability will not affect the outcome.
Best, a.y
From: <freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Linnea Marie Dramdal Borg <lmborg@stud.ntnu.nomailto:lmborg@stud.ntnu.no> Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Date: Monday, January 22, 2018 at 10:26 AM To: "freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Subject: [Freesurfer] TRACULA QC
Hi,
We have run into some issues with our tractographies. I have attached examples: In "fmajor_4053.png", there is a "bridge" in forceps minor on its way to the frontal lobe. We've also come across other cases of bridging in forceps minor, such as "fminor_4013". Also, there are some voxel paths that are rather weak, such as "unc_2068_1.png" and "unc_3022_1.png"/"unc_3022_2.png". This does not seem optimal. Do you agree?
We've also come across a couple of cases where it hasn't really found the tract correctly, such as in "29_ccg_1", "unc_3045_1.png" and "268_ccg.png". This also includes "fmm_32.png", which has a nice voxel intensity, but neither the forceps minor nor major voxels really fit their underlying tracts. Finally, neither "unc_3068_2.png" nor "unc_4040.png" are really correct.
Should we reinitialize all of them, or can we use some of them without affecting the results too much?
Thank you,
Linnea Marie