Hi Sabrina,
I talked to Bruce and he said we can improve the surfaces further if we
try labeling the lesion as "Left_Lesion" in the aseg and then running
recon-all from that point onward. It should freeze the white surface
exactly where you draw voxels (so be careful with it) but still let the
pial deform outwards. Right now, labeling the wm.mgz helps improve
things but the ?h.white surf still ignores the edits in some places. If
you directly edit the aseg then the surfaces will follow those
boundaries exactly.
I'll try it on a few slices of the data and let you know the results,
but you can also do it if you like.
-Ani
On 06/07/2016 04:46 PM, Sabrina Yu wrote:
Hi Ani,
I really appreciate the time and effort you are putting in to help me
find a solution to this problem. The surface does look much more
promising.
Following your instructions in the email, I have been playing around
with editing the wm.mgz and have been able to generate better wm
surfaces. With these edits, will I also be able to improve the
aparc.mgz and aparc+aseg.mgz?
Thanks so much,
Sabrina
------------------------------------------------------------------------
*From:* freesurfer-boun...@nmr.mgh.harvard.edu
<freesurfer-boun...@nmr.mgh.harvard.edu> on behalf of Ani Varjabedian
<a...@nmr.mgh.harvard.edu>
*Sent:* Tuesday, June 7, 2016 5:45:43 AM
*To:* freesurfer@nmr.mgh.harvard.edu
*Subject:* Re: [Freesurfer] Brainmask edit error
Hi Sabrina,
Just wanted to update you on my progress.
I tried editing the wm.mgz and filling in the hole to see if that
might help things. If we can get the wm surface accurate then there is
a good chance the pial surf (which grows from the white) will follow
suit.
After I edited the wm surf I ran this command to quickly generate the
next step and check prelim surfaces
recon-all -fill -tessellate -s SUBJECT01_wmedit
I made a copy of your subject to keep track of my edits, that's why
the name is a bit different. But basically I generated the filled
volume (from the wm.mgz) and then made the surfaces ?h.orig.nofix.
These are prelim surfaces to what will eventually become the wm
surfaces. If they look good then there is hope that this edit might do
the trick!
I've attached a screenshot and you can see that it did help push the
surfs out a bit more. I obviously did not edit the wm.mgz as carefully
or accurately as I could have but if this works then you can try it
yourself with more knowledge of the stroke region.
So! The next step is to see what the ?h.white surfs look like and to
do that I ran this command:
recon-all -smooth1 -inflate1 -qshpere -fix -white -s SUBJECT01_wmedit
I am getting these commands here:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReconAllDevTable
This is a really helpful table that lists the FS pipeline and all the
different steps so you can pick and choose which ones to run if you
want to see a result quickly.
I didn't check which version of FS you were using. I'm running the dev
version of 6.0 so some things might be a bit different but overall the
major steps should be the same enough to make a comparison.
To see a list of your recon-all steps you can always use the -dontrun
flag with your subject and it will just print out all the commands to
the logfile without actually running them. It's a really good way to
get a list of the steps and then you can look into what they do!
Anyways, when this is done I will update you on the white surfs. If
they look good then I will try the next step which will be generating
the pial!
Hope all is well and thank you again for being so patient during this
busy time.
-Ani
On 06/02/2016 02:15 PM, Ani Varjabedian wrote:
Hi Sabrina,
I'm not entirely sure about that, but if my manual edits don't work
I'll let Bruce know and we can discuss other workarounds.
-Ani
On 06/02/2016 12:55 PM, Sabrina Yu wrote:
Hi Ani,
Thanks so much for your explanation, it makes sense that the brain
is too deformed for one continuous surface. If we are not able to
find a workaround that creates a continuous surface, can I create a
separate surface just for the stroke region? And generate average
voxel intensity for that region?
Thanks,
Sabrina
On Jun 2, 2016, at 6:21 PM, Ani Varjabedian
<a...@nmr.mgh.harvard.edu> wrote:
Hi Sabrina,
Usually we don't add voxels in the brainmask. We usually add to the
wm.mgz which pushes the white surf out and ultimately the pial surf
out (because it grows from the white).
If the area around the stroke on both sides looks good but there is
a hole in the middle where there is no information or labels, then
the surfaces will not connect all the way around it. They treat it
like two separate pieces. Does this make sense? I'll try to play
around with some different things (like adding to the wm.mgz) and
see if I can get the surfaces to be continuous around this region.
Like Bruce said, it might end up being that the brain is too
different from what freesurfer assumes a brain to look like, but I
will try some work arounds and get back to you.
Thank you,
Ani
On 06/02/2016 08:18 AM, Sabrina Yu wrote:
Hi Ani,
Thanks for looking into this for me. I'm actually out of town
right now, so cannot access the subject in front of me.
Yes, it is. As I am trying to generate the stats on intensity over
an ROI, I am hoping to be able to generate good surfaces. To me,
the aseg looks pretty good. The brain matter around the stroke
area is recognized, but is not in the surfaces.
I only edited the brainmask.mgz. I unfortunately cannot tell you
the exact slices that I edited until I am back, but I tried to
recon edit and add in voxels in the stroke area. I did not notice
major problems with the surface including dura or skull, so I did
not really remove voxels. I'm not sure if how I did it is correct,
but I tried to follow the wiki page on pial surface edits. The
command I used to try to regenerate the surface was: recon-all
-subjid <SUBJECT> -autorecon2-pial
Thanks,
Sabrina
On Jun 1, 2016, at 4:39 PM, Ani Varjabedian
<a...@nmr.mgh.harvard.edu> wrote:
Hello Sabrina,
I have received the subject and am looking at it now.
I've attached a screenshot that is similar to the one you sent.
At [158, 115, 124] I see a hole in the aseg, wm mask, so it makes
sense that the brainmask does not cross over the area. Is this
the area you wish to include? I just want to make sure I'm
looking at the correct region.
What have you done so far to edit? I know you mentioned brainmask
editing, but what specifically did you do?
Thank you,
Ani
On 05/25/2016 08:47 AM, Ani Varjabedian wrote:
Hi Sabrina,
Could you upload the whole subject folder to the FTP or to our
filedrop? https://gate.nmr.mgh.harvard.edu/filedrop2/
I can play around with the data a bit to see if we can improve
surfaces. If that doesn't work then we can discuss other options
for how to get the information you need.
Thanks!
-Ani
On 05/23/2016 05:47 PM, Sabrina Yu wrote:
Thank you for your explanation. Would it be possible to find a workaround? I
want to generate stats on intensity over an ROI for these stroke subjects. For
the brain matter not included in the surface, can I create a volume label file
in freeview instead of trying to edit the surfaces?
Thank you,
Sabrina
________________________________________
From:freesurfer-boun...@nmr.mgh.harvard.edu <freesurfer-boun...@nmr.mgh.harvard.edu>
on behalf of Bruce Fischl<fis...@nmr.mgh.harvard.edu>
Sent: Monday, May 23, 2016 12:30:30 PM
To: Freesurfer support list
Subject: Re: [Freesurfer] Brainmask edit error
because it violates too many of our assumptions about the topology and
geometry of the brain.
cheers
Bruce
On Mon, 23 May 2016, Sabrina Yu
wrote:
Hi Bruce,
Thank you for your reply. I've uploaded the .mgz and surfaces using FTP to
FreeSurfer. Why is it that it may not be possible to get good surfaces with
extensive strokes?
Thank you,
Sabrina
________________________________________
From:freesurfer-boun...@nmr.mgh.harvard.edu <freesurfer-boun...@nmr.mgh.harvard.edu>
on behalf of Bruce Fischl<fis...@nmr.mgh.harvard.edu>
Sent: Thursday, May 19, 2016 11:24:51 AM
To: Freesurfer support list
Subject: Re: [Freesurfer] Brainmask edit error
Hi Sabrina
it's tough to tell from just the 2 slices. If you upload the subject one
of us will take a look, but it may not be possible to get good surfaces
with that extensive a stroke
cheers
Bruce
On Wed, 18 May 2016, Sabrina Yu wrote:
Hello,
I'm having problems with regenerating the surfaces for stroke cases after
editing the brainmask.mgz file. In the aseg.mgz file, the brain matter
around lesion area is correctly classified as brain matter. In the pial
surface, this brain matter is not included.
I've tried to edit the brainmask.mgz file to include this region, but when I
run: recon-all -subjid <SUBJECT> -autorecon2-pial, the surfaces generated
are not changed at all.
I have attached some images to this email. I'm new to Freesurfer, so any
help would be greatly appreciated.
Thank you.
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
<SUBJECT01.png>
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
<mailto:Freesurfer@nmr.mgh.harvard.edu>
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to
whom it is
addressed. If you believe this e-mail was sent to you in error
and the e-mail
contains patient information, please contact the Partners
Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent
to you in error
but does not contain patient information, please contact the
sender and properly
dispose of the e-mail.
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu>
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to
whom it is
addressed. If you believe this e-mail was sent to you in error and
the e-mail
contains patient information, please contact the Partners
Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to
you in error
but does not contain patient information, please contact the sender
and properly
dispose of the e-mail.
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
