Dear Bruce,
Thank you for your help. I was able to view the defects. It does seem to be an extensive defect—possibly retained dura? I’m not sure of any other way to fix it. Do you think that the poor quality of the source image makes it impossible to generate any kind of surface using free surfer?
Best, Dan
On May 11, 2017, at 12:00 PM, freesurfer-request@nmr.mgh.harvard.edumailto:freesurfer-request@nmr.mgh.harvard.edu wrote:
Message: 9 Date: Thu, 11 May 2017 11:38:42 -0400 (EDT) From: Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> Subject: Re: [Freesurfer] Freesurfer Digest, Vol 159, Issue 12 To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Message-ID: <alpine.LRH.2.20.1705111132360.28473@door.nmr.mgh.harvard.edumailto:alpine.LRH.2.20.1705111132360.28473@door.nmr.mgh.harvard.edu> Content-Type: text/plain; charset="iso-8859-7"
Hi Dan
that's a pretty low-contrast, severely motion-corrupted scan. We did pretty well given the input I think. To visualize the defects you can cd into the subject's mri dir and run:
fv -v brain.mgz wm.mgz:colormap=heat -f \ ../surf/lh.orig.nofix:overlay=../surf/lh.defect_labels:overlay_threshold=1,100 \ ../surf/lh.inflated.nofix:overlay=../surf/lh.defect_labels:overlay_threshold=1,100
then click the "use overlay color" button for the lh.orig.nofix surface. You can also change the display paramters in "configure overlay" to find a specific defect by setting the min/max to a narrow range around the one you are interested in (this will put it in the middle of the range and color it orange)
cheers Bruce
On Thu, 11 May 2017, Weisholtz, Daniel S.,M.D. wrote:
Dear Bruce, Thank you so much for your suggestions. Unfortunately, I was not able to open lh.defect_labels in freeview. ?When I try, Freeview crashes. ?There are no rh surfaces in the surf folder, so I presume this means the problem occurred in the left hemisphere. I was able to view the inflated.nofix and orig.nofix but I?m still not exactly sure what the problem is, and I would be grateful if you could take a look at the subject folder. I will upload it to the ftp site. It is called s117.
Thank you!
Best, Dan
On May 10, 2017, at 10:43 AM, freesurfer-request@nmr.mgh.harvard.edumailto:freesurfer-request@nmr.mgh.harvard.edu wrote:
Message: 9 Date: Tue, 9 May 2017 15:54:12 -0400 (EDT) From: Bruce Fischl <fischl@nmr.mgh.harvard.edumailto:fischl@nmr.mgh.harvard.edu> Subject: Re: [Freesurfer] recon-all freezes while CORRECTING DEFECT 60 To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu> Message-ID: <alpine.LRH.2.20.1705091553090.16219@door.nmr.mgh.harvard.edumailto:alpine.LRH.2.20.1705091553090.16219@door.nmr.mgh.harvard.edu> Content-Type: text/plain; charset="iso-8859-15"
Hi Daniel
you can load the ?h.defect_labels as an overlay on either the? inflated.nofix or orig.nofix surface and use it to find the location of? that defect. Usually this means something big is wrong - it should be? obvious looking at the inflated.nofix I would think. If you can't find it,? upload the subject to our ftp site and I'll take a look Bruce
On Tue, 9 May 2017, Weisholtz, Daniel S.,M.D. wrote:
Hello Freesurfer Developers, I am trying to run recon-all -all and the procedure is freezing while trying to correct what appears to be a large defect (vertices=36706). The source T1 image is not great quality due to rippling artifact, but I have inspected brainmask.mgz, wm.mgz, and aseg.mgz and I cannot identify a large defect. The dura was not completely removed, and I tried improving this by using various watershed thresholds as well as using gcut. I also tried manually removing the remaining dura, but none of this was helpful. ?I also tried using -T2FLAIR, but this did not seem helpful either.?
I am attaching the recon-all.log. This does not include the edits I made to brainmask.mgz. That was done on a separate attempt. This does include the -T2FLAIR, as you can see in the recon-all.log file.
1) FreeSurfer version:?freesurfer-Darwin-lion-stable-pub-v5.3.0 2) Platform: OS X Yosemite 10.10.5 3) recon-all.log attached
I very much appreciate any assistance you can provide. Happy to send any of the data if it would be helpful.
Sincerely, Dan Weisholtz