what does the aseg look like? Why isn't the cerebllum being removed? Dura is harder (unless you have a T2 or FLAIR) On Sun, 10 May 2020, Billah, Tashrif wrote:

Running the commands separately, I see the following log for some of the defects. Is the weblink my cue?

CORRECTING DEFECT 0 (vertices=4912, convex hull=886, v0=0) An extra large defect has been detected... This often happens because cerebellum or dura has not been removed from wm.mgz. This may cause recon-all to run very slowly or crash. if so, seehttps://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TopologicalDefect_free view

-Tashrif .

---

From: Billah, Tashrif Sent: Sunday, May 10, 2020 2:11 PM To: freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu Cc: Bouix, Sylvain,Ph.D. SYLVAIN@BWH.HARVARD.EDU; Knyazhanskaya, Evdokiya eknyazhanskaya@bwh.harvard.edu Subject: mris_fix_topology command does not finish   Hi there,

The following commands part of `recon-alll -i some_T1w.nii.gz -autorecon2` does not finish for some of our data:

#@# Fix Topology lh Sat May  9 14:17:52 EDT 2020 mris_fix_topology -mgz -sphere qsphere.nofix -inflated inflated.nofix -orig orig.nofix -out orig.premesh -ga -seed 1234 sub-7049_ses-01_desc-XcMaN4_T1w lh

#@# Fix Topology rh Sat May  9 14:17:52 EDT 2020 mris_fix_topology -mgz -sphere qsphere.nofix -inflated inflated.nofix -orig orig.nofix -out orig.premesh -ga -seed 1234 sub-7049_ses-01_desc-XcMaN4_T1w rh

Waiting for PID 57296 of (57296 57299) to complete... Waiting for PID 57299 of (57296 57299) to complete...

We kept them running for about 48 hours in multiple attempts, yet no luck. Moreover, mris_fix_topology​ is a binary file so I am unable to debug through.

Any suggestion on where to look at?

Best, Tashrif

no, I wouldn't think so. YOu can try mris_topo_fixer instead if you want, but really you need to diagnose what the cause of your problem is

On Sun, 10 May 2020, Billah, Tashrif wrote:

Hi Bruce,

Thanks for your prompt reply. I shall look at them as you noted. In the meantime, putting in some more logs from the perpetuating defect correction by mris_fix_topology:

13381 modVnum: vno:23308 has 6673 immediate neighbours mrisurf_topology.cpp:370 13382 vno:23308 now has 6673 immediate neighbours  mrisurf_topology.cpp:400 13383 modVnum: vno:23308 has 6674 immediate neighbours mrisurf_topology.cpp:370 13384 vno:23308 now has 6674 immediate neighbours  mrisurf_topology.cpp:400 13385 modVnum: vno:23308 has 6675 immediate neighbours mrisurf_topology.cpp:370 13386 vno:23308 now has 6675 immediate neighbours  mrisurf_topology.cpp:400 13387 modVnum: vno:23308 has 6676 immediate neighbours mrisurf_topology.cpp:370 13388 vno:23308 now has 6676 immediate neighbours  mrisurf_topology.cpp:400 13389 modVnum: vno:23308 has 6677 immediate neighbours mrisurf_topology.cpp:370 13390 vno:23308 now has 6677 immediate neighbours  mrisurf_topology.cpp:400 13391 modVnum: vno:23308 has 6678 immediate neighbours mrisurf_topology.cpp:370 13392 vno:23308 now has 6678 immediate neighbours  mrisurf_topology.cpp:400

However, will it help to play with the following args of mris_fix_topology?

-niters N                 stop genetic algorithm after N iterations

-seed N                 set random number generator to seed N

-s N                 smooth corrected surface by N iterations

Cheers, Tashrif

---

From: Billah, Tashrif tbillah@bwh.harvard.edu Sent: Sunday, May 10, 2020 3:05 PM To: freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu Subject: Re: mris_fix_topology command does not finish   Running the commands separately, I see the following log for some of the defects. Is the weblink my cue?

CORRECTING DEFECT 0 (vertices=4912, convex hull=886, v0=0) An extra large defect has been detected... This often happens because cerebellum or dura has not been removed from wm.mgz. This may cause recon-all to run very slowly or crash. if so, seehttps://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/TopologicalDefect_free view

-Tashrif .

---

From: Billah, Tashrif Sent: Sunday, May 10, 2020 2:11 PM To: freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu Cc: Bouix, Sylvain,Ph.D. SYLVAIN@BWH.HARVARD.EDU; Knyazhanskaya, Evdokiya eknyazhanskaya@bwh.harvard.edu Subject: mris_fix_topology command does not finish   Hi there,

The following commands part of `recon-alll -i some_T1w.nii.gz -autorecon2` does not finish for some of our data:

#@# Fix Topology lh Sat May  9 14:17:52 EDT 2020 mris_fix_topology -mgz -sphere qsphere.nofix -inflated inflated.nofix -orig orig.nofix -out orig.premesh -ga -seed 1234 sub-7049_ses-01_desc-XcMaN4_T1w lh

#@# Fix Topology rh Sat May  9 14:17:52 EDT 2020 mris_fix_topology -mgz -sphere qsphere.nofix -inflated inflated.nofix -orig orig.nofix -out orig.premesh -ga -seed 1234 sub-7049_ses-01_desc-XcMaN4_T1w rh

Waiting for PID 57296 of (57296 57299) to complete... Waiting for PID 57299 of (57296 57299) to complete...

We kept them running for about 48 hours in multiple attempts, yet no luck. Moreover, mris_fix_topology​ is a binary file so I am unable to debug through.

Any suggestion on where to look at?

Best, Tashrif