External Email - Use Caution

Hi Marco, Just added to the recon-all script on dev branch the read of T2 min/max inside/outside and nsigma_below/above parameters, which are now being passed to mris_make_surfaces. As I mentioned on the github bug you created, I would be careful playing with those parameters since the T2 image has been normalized for intensity and those default parameters should be pretty close.

But let me know how it goes! Best!

Viv

On Thu, Aug 22, 2019 at 8:29 AM Marco Aqil m.aqil@spinozacentre.nl wrote:

    External Email - Use Caution

Hi,

I would like to report a bug in freesurfer-dev. I have encountered this bug in a version downloaded on July 11th, 2019, and again in a version downloaded two days ago (August 20th). In short, it seems that parameters specified via an expert file relevant to T2-assisted pial segmentation, such as -T2_min, are not passed to the second call of mris_make_surfaces (the one actually using T2 data).

In Freesurfer 6.0.0, this is not the case, and the T2_min parameter is indeed passed to the second call of mris_make_surfaces (although I note that in 6.0.0, when specifying -nsigma_below via the expert file, both the user-specified value and the standard value are passed to the second call of mris_make_surfaces). See below for more details on the specific freesurfer-dev behavior.

Best,

Marco Aqil

*Bug description: T2-specific parameters not passed to second call of mri_make_surfaces in freesurfer-dev*

*Call: *recon-all autorecon3​ -subjid <subject id> -hires -T2pial -parallel -openmp 15 -expert <expert file path>

*Expert file content: *mris_make_surfaces -T2_min 35

...

From the recon-all log, the first call of mris_make_surfaces receives the

-T2_min parameter:

---

*#@# Make Pial Surf lh Mon Aug 19 13:43:02 CEST 2019*

*<scripts folder path>*

*mris_make_surfaces -save-res -save-target -orig_white white.preaparc -orig_pial white.preaparc -aseg ../mri/aseg.presurf -mgz -T1 brain.finalsurfs -T2_min 35 sub-006 lh*

*----------------*

However, the second call of mris_make_surfaces, which relies on T2 data to refine the pial, does not receive the T2_min parameter, and looks as follows:

---

*#@# Refine Pial Surfs w/ T2/FLAIR Mon Aug 19 14:02:23 CEST 2019* *<other T2 pial steps>*

*mris_make_surfaces -save-res -save-target -orig_white white -orig_pial woT2.pial -aseg ../mri/aseg.presurf -nowhite -mgz -T1 brain.finalsurfs -T2 ../mri/T2 -nsigma_above 2 -nsigma_below 5 sub-006 lh *

*---------------*

I observed the same behavior when specifying -nsigma_below via the expert file.

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer