We will give 7.2 a try as well.

Lilla’s suggestion did create/append the CC to the aseg using:

set cmd = (mri_cc -aseg aseg.mgz -o aseg_CCseg.mgz $subj)

to get the CC lables.

Thx.

Jim

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of Yendiki, Anastasia
Sent: Thursday, October 21, 2021 6:56 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Using only a aparc+aseg input file for TRACULA

* External Email - Caution *

This won't be good for the cc.* tracts for sure (in the new version we have the entire CC covered with 7 subdivisions). TRACULA learns how often the n-th point along tract X falls in or next to label Y from the structural segmentation. If it's learned that a tract goes through label Y=254 half way along its trajectory, and where our segmentation has label 254 yours has something else, it won't a great situation. It could still work though, because other terms of the objective function will drive the fitting (the likelihood term that fits the shape of the tracts to the diffusion orientations, or the anatomical priors in other points along the tract). In general, I'm still surprised by how robust TRACULA turns out to be to various types of snafus. So it's worth trying.

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Alexopoulos, Dimitrios <dimitriosalexopoulos@wustl.edu>
Sent: Thursday, October 21, 2021 2:59 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Using only a aparc+aseg input file for TRACULA

External Email - Use Caution

Anastasia,

Upon closer inspection the aparc+aseg we have are analogous to the FS labeling, however, the CC is not segmented, it is simply labelled as cortical WM (labels 2/41). Will using Tracula even work at all or will it generate the other tracts but not include the CC minor/major tracts?

Jim

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of Yendiki, Anastasia
Sent: Wednesday, October 20, 2021 2:12 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Using only a aparc+aseg input file for TRACULA

* External Email - Caution *

Hi Jim - If the file was called $SUBJECTS_DIR/$subj/mri/aparc+aseg.mgz, included the same labels and followed the same lookup table as the OG aparc+aseg, this could work. Note that, for the new TRACULA (in FS 7.2), we recommend also running the thalamic nuclei segmentation, as it helps with the accuracy of the tracts that terminate or neighbor the thalamus.

Anastasia.

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Alexopoulos, Dimitrios <dimitriosalexopoulos@wustl.edu>
Sent: Wednesday, October 20, 2021 2:54 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Using only a aparc+aseg input file for TRACULA

External Email - Use Caution

Hi,

It was my understanding that TRACULA only uses the aparc+aseg segmentation from the standard structural output from FS 7.1.1., and specifically for the relative positions (left, right, anterior, etc.) of each tract with respect to the segmentation labels in the aparc+aseg.

Is this correct?

How would we run trac-all if we only had an aprc+aseg from another pipeline that generates FS-compatible outputs?

For example, trac-all -preproc -bedp -path -c (where c = dmrirc config file).

Jim

The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.