Freesurfer

freesurfer@nmr.mgh.harvard.edu

15 participants
30064 discussions

Parcellation and bias field correction
by Mike Chee 19 Jul '06

19 Jul '06

----- Forwarded Message From: Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> Date: Wed, 19 Jul 2006 08:42:08 -0400 (EDT) To: Mike Chee <mchee(a)pacific.net.sg> Hi Mike A. the wm parcellation isn't released yet, but I think we'll do so soon (Doug: what do you think). The current tools should allow you to correlate demographic/neuropsych with parcellation units (e.g. superior frontal gyrus). B. The improved bias correction should help both the cortical and subcortical streams. C. The control points range of effect depends on the other control points (auto and manual) near it). If a control point is anterior to all others it can have an effect on the entire frontal region. you should post these to the list - other people can answer some of them (like A) better than I can. cheers, Bruce On Wed, 19 Jul 2006, Mike Chee wrote: > Hi Bruce, > > I have a couple of questions: > A. We are interested in correlating parcellated brain volumes of brain lobes > (not subcortical structures for openers) with neuropsych test scores. For > example, middle frontal gyrus volume with executive function. Presently, it > seems that it is possible to evaluate cortical thickness globally and then > pick our areas that regress to a significant extent with variables of > interest. At some point in time, it would be good to see if the combination > of GM + the underlying WM would give additional / complimentary information. > In this regard, I am not entirely clear how we are to realize this goal. > According to David’s HBM 2006 poster, this would take place by adding the > volume of the cortical rim and its underlying WM. > B. If so, this would necessitate good quality segmentation along the > volumetric pipeline (that generates the .aseg images). This in turn would > require some adjustments in the segmentation of white matter using the > volume processing stream. By dropping control points, we active improved > segmentation along the surface processing stream (correct?) If so, how are > benefits to volume measurement realized? As I understand our data, the > surface processing pipeline is more or less satisfactory as is. So I wonder > what the marginal benefit of dropping control points is? > C. A final question relates to how laying control points in WM works. Jenni > told us that dropping more points increases intensity in an additive > fashion. If so what is the intensity shift and how extensive is the induced > change? > If you think these questions serve a general audience, feel free to post > this question in the help website. > > Mike > ------ End of Forwarded Message

1 0

Expanding pial outline
by Jake Freimer 18 Jul '06

18 Jul '06

I ran two brains through autorecon2 and the pial line is drawn too close at many parts and cuts off parts of the brain. I looked through the website and only saw instructions for how to remove stuff and make the pial line smaller. I was wondering if there is a command I can run to expand the pial line or what is the best way to edit the files so more brain is included. I was looking at the threshold tools and couldn't see a way to convert say anything with a value of 30-40 to 60, so I was wondering if that was also possible. Thanks -- -Jake

2 1

viewing subcortical structures
by Sasha Wolosin 18 Jul '06

18 Jul '06

Dear all, Does aseg save out masks of the various subcortical regions measured? If so, is it possible to visualize and/or edit these in tkmedit/scuba, or export them to an external format? Thanks, Sasha Disclaimer: The materials in this e-mail are private and may contain Protected Health Information. Please note that e-mail is not necessarily confidential or secure. Your use of e-mail constitutes your acknowledgment of these confidentiality and security limitations. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying, distribution, or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender via telephone or return e-mail.

3 5

Re: [Freesurfer] recon-all error
by Anil Roy 17 Jul '06

17 Jul '06

That could be the case. I start recon-all and just leave the computer for 24 hrs. Is there a way to find out what could be doing this. Anil. On 7/17/06, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> wrote: > > is there some builtin time limit that you are exceeding? The log file just > says: > > mri_remove_neck -radius 25 nu.mgz transforms/talairach.m3z > /usr/local/freesurf\ > er/average/RB_all_2006-02-15.gca nu_noneck.mgz > > erasing everything more than 25 mm from possible brain > reading atlas '/usr/local/freesurfer/average/RB_all_2006-02-15.gca'... > setting gca type = Normal gca type > reading input volume 'nu.mgz'... > reading transform 'transforms/talairach.m3z'... > Killed > Linux dhcp-129-105-146-221.speech.northwestern.edu 2.4.22-10mdk #1 Thu Sep > 18 1\ > 2:30:58 CEST 2003 i686 unknown unknown GNU/Linux > > recon-all exited with ERRORS at Sat Jul 15 01:11:14 EDT 2006 > > > > which sounds like some other process killed this one. > > Bruc > > On Mon, 17 Jul > 2006, Anil Roy wrote: > > > Hello All, > > > > I've been running recon-all on a couple of minc brains on Fedora Core 4 > and > > everything thing runs fine. I started running a few on a Mandrake > > Linux 9.2system but each time recon-all exits with errors and I'm > > unable to see why. > > I've attached the recon-all.log file and would appreciate any input on > why > > this is occurirng. > > > > Thanks, > > > > Anil Roy. > > >

1 0

Hemispheric devision problem
by Akram Bakkour 17 Jul '06

17 Jul '06

Hi all, I've been having trouble with these two brains with huge ventricles, first the aseg failed, and the ventricles didn't get filled automatically, so I filled them manually and ran recon-all -autorecon2-wm but my surfaces were off and there was a blatant hemispheric devision problem with the surfaces, so I added a corpus callosum seed point using recon-all -autorecon2-wm -cc-crs <X Y Z> but that didn't work. I ran out of fixes to try, any suggestions!? Thanks! On machine blakey source /usr/local/freesurfer/nmr-std-env setenv SUBJECTS_DIR /space/banzai/3/users/efenster/CLINDEM_processing/ which mri_convert /usr/local/freesurfer/stable/bin//mri_convert cd $SUBJECTS_DIR recon-all -autorecon2-wm -cc-crs 126 98 97 -subjid 06142006_4TTclindem03 subjid 06142006_4TTclindem03 recon-all -autorecon2-wm -cc-crs 129 113 128 -subjid 06212006_4TTclindem04 subjid 06212006_4TTclindem04 check output: foreach s (06142006_4TTclindem03 06212006_4TTclindem04) tkmedit 06142006_4TTclindem03 lh.white -aux wm.mgz -aux-surface rh.white end Notice the hemispheric devision problem with the surfaces. Any suggestions as to how to fix this!? Thanks! Akram.

2 1

Question concerning orientation
by Jake Freimer 17 Jul '06

17 Jul '06

I have run several instances of autorecon2 and have only got left hemisphere data. The log file says that autorecon finished with errors but I can not find where to find these errors. I believe that the problem is the way the data being fed in is oriented so I was wondering how I should reorient it to get it to work. Thanks Jake

3 2

general question concerning orientation
by Chacko Cherian 17 Jul '06

17 Jul '06

Hello everyone, orientation i guess is quite an issue. Well how does,Freesurfer determine orientation of the image.Do I have to know how the image has been processed before it reaches me or is there any other way to find out?Does it have to do with the 'flip'? I checked the orientation of an image on SPM and on Freesurfer. Left on SPM is Right on Freesurfer and vice versa is what I got. Essentially how do I determine L-R orientation? Thanks in advance, Chacko Cherian. Arizona State University. __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

recon-all error
by Anil Roy 17 Jul '06

17 Jul '06

Hello All, I've been running recon-all on a couple of minc brains on Fedora Core 4 and everything thing runs fine. I started running a few on a Mandrake Linux 9.2system but each time recon-all exits with errors and I'm unable to see why. I've attached the recon-all.log file and would appreciate any input on why this is occurirng. Thanks, Anil Roy.

1 0

problems running FreeSurfer on a Scyld BeoWulf cluster
by Renat Yakupov 17 Jul '06

17 Jul '06

Hello everybody, Does anybody run FreeSurfer on Linux clusters in general or Scyld BeoWulf clusters in particular? Do you have any problems? We are having some problems running FreeSurfer on our cluster. Quite often when we run autorecon2 stage on a compute node, it crashes. Sometimes a node crashes very quickly, sometimes after hours of processing. All of the 4 compute nodes that we have here have crashed. And it is not hardware related, because I ran linpack on them for 50+ hours in a row, and it was fine. Any ideas what might be causing it? Any suggestions? We run freesurfer-Linux-centos4_x86_64-stable-pub-v3.0.3-full.tar.gz <http://surfer.nmr.mgh.harvard.edu/pub/dist/freesurfer-Linux-centos4_x86_64- stable-pub-v3.0.3-full.tar.gz> version, and here is `uname -a` output: Linux scyld.localdomain 2.4.21-37.EL.301r0072smp #1 SMP Wed Mar 1 04:09:49 EST 2006 x86_64 x86_64 x86_64 GNU/Linux Thank you. Regards, Renat.

2 1

OpenGL
by Joe Berens 17 Jul '06

17 Jul '06

Hello, A user gets the following: GLUT: Fatal Error in tkmedit: OpenGL GLX extension not supported by display: <hostname>:60.0 This is using tightvnc via ssh from a windows xp box to Redhat Enterprise 3. Every thing else seems to work fine. Any Ideas? -Joe A __________________________________________________________________________ Joe Berens Department of Electrical and Computer Engineering University of Minnesota - Twin Cities Phone: 612-625-4585 200 Union Street SE, Room 4-174 Email: joe(a)umn.edu Minneapolis, MN 55455 Office: EE/CSsci 1-183 Department of Psychology University of Minnesota - Twin Cities Phone: 612-625-0644 75 East River Road Email: joe(a)umn.edu Minneapolis, MN 55455 Office: N219 Elliot Hall __________________________________________________________________________

3 3

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer