Freesurfer June 2004

freesurfer@nmr.mgh.harvard.edu

24 participants
25 discussions

Parcellate_subject
by Chekema Nayasha Prince 30 Jun '04

30 Jun '04

I ran parcellate subject and I overlayed the annotation on the inflated surface. The medial wall segmentation looks incorrect. The medial wall area section is very large and is also spreading into the frontal regions. I looked at the T1 image with the lh.white surface loaded in tkmedit and the outline around the wm and gm look ok. This is the first time I have seen the medial wall appear so wide spread in the parcellated inflate brain. Is there a way to either fix the problem or investigate what is causing it? -Keyma

2 1

non-primate brains
by Geoffrey Aguirre 29 Jun '04

29 Jun '04

Free surfers -- I am collecting data from dogs. Any advice regarding passing this data through the system? I have a couple of specific questions: 1) Should I acquire data with voxel sizes smaller than 1mm, given that the brain is much smaller? How would I tell free-surfer to operate with this higher resolution data? 2) So far, automated skull stripping has been a failure, even with definition of a "seed point" in the white matter. The routine strips away the brain around the seed point, leaving a large chunk of scalp and muscle with a bit of cortex attached. Any thoughts on how to get this going? Thanks in advance for any help, Geoff -- Geoffrey Karl Aguirre, M.D., Ph.D. University of Pennsylvania Center for Cognitive Neuroscience 3815 Walnut Street Fax: (215) 898-1982 Philadelphia, PA 19104-6196 mailto:aguirre@neuro.med.upenn.edu http://ccn.upenn.edu/~aguirre

8 10

segmentation fault with mprph_subject
by Chekema Nayasha Prince 29 Jun '04

29 Jun '04

Hello, I received the following segmentation fault when I ran morph_subject: mris2rgb -o hr_MPRAGE_040623/.reg -both -c /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//surf/lh.sulc -canon /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//surf/lh.sphere.reg /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//surf/lh.sphere.reg /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//rgb ------------------------------------- using hr_MPRAGE_040623/.reg as output stem reading canonical coordinates from /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//surf/lh.sphere.reg. writing rgb file /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//rgb/lateral.lh.hr_MPRAGE_040623/.reg.rgb iopen: can't open output file /home/freesurfer/freesurfer-20040218/freesurfer//subjects/hr_MPRAGE_040623//rgb/lateral.lh.hr_MPRAGE_040623/.reg.rgb No such file or directory Segmentation fault mris2rgb exited with errors After checking the message thread I noticed that someone else also had a similar error and you recommended that they comment out the mris2rgb line. I looked in the morph_rgb-lh and rh files and the call to mris2rgb was already commented out. Is there any other other reason that this error could be appearing. I ran the subject with recon-all -stage1 and then I did the manual edits and run fix_subject and make_final_surfaces_subject. thanks -keyma

2 1

Freesurfer on Fedora Core 2
by N.M. van Strien 25 Jun '04

25 Jun '04

Dear All, With the help of the people at freesurfer I managed to get freesurfer working on the newly available distribution of Fedora Core 2. This mail covers an account of how we managed to get things working, so it is less informative to people who are not planning to use Fedora Core 2. My system hardware: CPU: Dual Intel Xeon 2.4 Ghz (HT disabled) Video: Nvidia Geforce 4 TI 4600 RAM: 2 GB Required Software: Fedora core 2 -> can be downloaded here http://fedora.redhat.com/download/mirrors.html Fedora kernel compiled with 8 stacks -> http://www.linuxant.com/driverloader/wlan/full/downloads-fc2-kernel-i686.php Nvidia Linux driver v53.16 -> http://www.nvidia.com/object/linux.html Freesurfer -> version of February 18th 2004 http://surfer.nmr.mgh.harvard.edu/download.html TCL/TK 8.3.5 -> sources http://www.tcl.tk/software/tcltk/downloadnow83.html Overview of the installation: Step 0: Download all the required packages Step 1: Install Fedora Step 2: Install Fedora kernel compiled with 8 stacks (required for compatibility with the Nvidia driver) Step 3: Install Nvidia driver Step 4: Install Freesurfer Step 5: Compile and install TCL/TK Step 6: Let's play with Freesurfer Installation details: Step 0: Make sure to have all the packages before starting the installation. If you already have Fedora Core running on your system, download the remaining packages and start at Step 2. Step 1: Installation instructions for downloading and installing Fedora Core 2 are covered on the fedora project website. (www.fedora.redhat.com <http://www.fedora.redhat.com/> ). Make sure to use the tc or c shell. You can set your default shell in /etc/passwd. Fedora Core 2 runs bash by default. Step 2: Download the kernel compiled with 8 stacks or compile it yourself. Both the normal and smp (multi processor) core are available from the website above. Installation of the rpm is simple. First unzip and then rpm -i kernel-smp-2.6.6-1.427.8kstacks.i686.rpm Next, reboot into the 8kstacks kernel and start with runlevel 3. You can set the default runlevel in /etc/inittab. Step 3: Installation of the Nvidia driver can only be performed from the command line. The X server should not be running. Installation instructions for downloading and installing the driver can be found here ftp://download.nvidia.com/XFree86/Linux-x86/1.0-5336/README. Once installed, check to see if your OPENGL GLX driver is working properly. Type at the command line: glxinfo Check that direct rendering is set to YES Next, type glxgears Can you see the gears rotating? How many frames do you get? If this is all working, you are ready for the next step. Step 4: Install Freesurfer. Installation instructions for downloading and installing Freesurfer can be found here ftp://surfer.nmr.mgh.harvard.edu/pub/docs/INSTALL. Make sure to correctly setup the shell and all variables, as is outlined in the installation documentation. To check if tkmedit and tksurfer will make proper use of the nvidia driver, type ldd /usr/local/bin/freesurfer/bin/Linux/tkmedit and also ldd /usr/local/bin/freesurfer/bin/Linux/tksurfer Note the files being used for libGL and libGLcore -- if they are something other than the NVIDIA libraries, then you will need to either remove the libraries that are getting in the way, or adjust your ld search path. (See the Nvidia driver installation manual.) Step 5: Fedora Core 2 comes with the latest release of TCL/TK (=8.4.X). Since tkmedit and tk surfer are compiled with the 8.3.X version, an incompatibility exists and tkmedit and tksurfer will not start properly. In order to overcome this problem you will have to compile TCL/TK 8.3.X for your system and point tkmedit and tksurfer to the right version. You can learn how to do this here http://www.tcl.tk/doc/howto/compile.html. Since there are no 8.3.X binaries compiled for Fedora Core 2, I see no other option than compiling your own. This is relatively simple by typing the commands below in the /usr/src/tcl8.3.5/unix source path (or wherever you unpacked them). The same goes for tk8.3.X ./configure make make test make install Finally I created a symlink in /lib pointing to /usr/lib/tcl8.3.x by typing ln -s -i /usr/lib/tcl8.3.x/libtcl8.3.so.0 /lib/ libtcl8.3.so.0 and the same for tk ln -s -I /usr/lib/tk8.3.x/libtk8.3.so.0 /lib/libtk8.3.so.0 Step 6: Happy playing. drs. Niels M. van Strien Vrije Universiteit Medisch Centrum Department of Anatomy, room MF-G102 b Van der Boechorststraat 7 P.O. Box 7057 1007 MB Amsterdam The Netherlands e-mail: n.vanstrien(a)vumc.nl phone: +31 (20) 4448047 fax: +31 (20)

1 0

SNR values
by Chekema Nayasha Prince 24 Jun '04

24 Jun '04

Is it possible to get SNR values from freesurfer for regions of the brain? -Keyma

2 1

missing curvatures
by nima＠nmr.mgh.harvard.edu 24 Jun '04

24 Jun '04

hi after "making final surfaces" i enountered something strange! lh: "inflated" brain and "smoothwm" has no curvature but other surfaces like orig va qshere are normal rh: inflated brain has the curvature but other surfaces are normal logfiles are shown below. what is the prob?!?!? any idea? cheers nimA FOR LEFT HEMISPHERE "INFLATED": " env tksurferinterface=csurf patchname= /space/lyon/1/home/inverse/freesurfer_alpha/bin/Linux/tksurfer -sleep_s2 lh inflated -tcl /tmp/TmpTclSurf.12210 MRISreadVertexPositions(/space/humpback/2/users/nima/csurf/sleep_s2/surf/lhorig): surfaces differ. No such file or directory MRISreadOriginalProperties: could not read surface file /space/humpback/2/users/nima/csurf/sleep_s2/surf/lh.orig No such file or directory Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkm_wrappers.tcl Using libtclfsgdf.so Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Successfully parsed tksurfer.tcl MRISreadNewCurvature: incompatible vertex number in file /space/humpback/2/users/nima/csurf/sleep_s2/surf/lh.curv surfer: current subjects dir: /space/humpback/2/users/nima/csurf surfer: in subjects "scripts" dir surfer: session root data dir ($session) set to: surfer: /autofs/space/humpback_002/users/nima/csurf/sleep_s2 surfer: /space/humpback/2/users/nima/csurf/sleep_s2/mri/T1/COR-.info surfer: vertices=125362, faces=250720 surfer: using interface /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tksurfer.new.tcl Read /space/lyon/1/home/inverse/freesurfer_alpha/surface_labels.txt, found 95 structures tksurfer: run tcl script: /tmp/TmpTclSurf.12210 surfer: single buffered window surfer: curvature read: min=0.000000 max=0.000000 " ps. point is that lh.orig exists! so i do not know what it says that it can not find it! FOR LEFT HEMISPHERE "INFLATED": "env tksurferinterface=csurf patchname= /space/lyon/1/home/inverse/freesurfer_alpha/bin/Linux/tksurfer -sleep_s2 lh smoothwm -tcl /tmp/TmpTclSurf.12210 MRISreadVertexPositions(/space/humpback/2/users/nima/csurf/sleep_s2/surf/lhorig): surfaces differ. No such file or directory MRISreadOriginalProperties: could not read surface file /space/humpback/2/users/nima/csurf/sleep_s2/surf/lh.orig No such file or directory Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkm_wrappers.tcl Using libtclfsgdf.so Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Successfully parsed tksurfer.tcl MRISreadNewCurvature: incompatible vertex number in file /space/humpback/2/users/nima/csurf/sleep_s2/surf/lh.curv surfer: current subjects dir: /space/humpback/2/users/nima/csurf surfer: in subjects "scripts" dir surfer: session root data dir ($session) set to: surfer: /autofs/space/humpback_002/users/nima/csurf/sleep_s2 surfer: /space/humpback/2/users/nima/csurf/sleep_s2/mri/T1/COR-.info surfer: vertices=125362, faces=250720 surfer: using interface /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tksurfer.new.tcl Read /space/lyon/1/home/inverse/freesurfer_alpha/surface_labels.txt, found 95 structures tksurfer: run tcl script: /tmp/TmpTclSurf.12210 surfer: single buffered window surfer: curvature read: min=0.000000 max=0.000000" ps. point is that lh.orig exists! so i do not know what it says that it can not find it! FOR RIGHT HEMISPHERE "INFLATED SURFACE": ha/bin/Linux/tksurfer -sleep_s2 rh qsphere -tcl /tmp/TmpTclSurf.12210 mrisReadTriangleFile(/space/humpback/2/users/nima/csurf/sleep_s2/surf/rh.orig): surface doesn't match /space/humpback/2/users/nima/csurf/sleep_s2/surf/rh.qsphere No such file or directory mrisReadTriangleFile failed. No such file or directory MRISreadOriginalProperties: could not read surface file /space/humpback/2/users/nima/csurf/sleep_s2/surf/rh.orig No such file or directory Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkm_wrappers.tcl Using libtclfsgdf.so Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Reading /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tkUtils.tcl Successfully parsed tksurfer.tcl MRISreadNewCurvature: incompatible vertex number in file /space/humpback/2/users/nima/csurf/sleep_s2/surf/rh.curv surfer: current subjects dir: /space/humpback/2/users/nima/csurf surfer: in subjects "scripts" dir surfer: session root data dir ($session) set to: surfer: /autofs/space/humpback_002/users/nima/csurf/sleep_s2 surfer: /space/humpback/2/users/nima/csurf/sleep_s2/mri/T1/COR-.info surfer: vertices=127844, faces=255756 surfer: using interface /space/lyon/1/home/inverse/freesurfer_alpha/lib/tcl/tksurfer.new.tcl Read /space/lyon/1/home/inverse/freesurfer_alpha/surface_labels.txt, found 95 structures tksurfer: run tcl script: /tmp/TmpTclSurf.12210 surfer: single buffered window surfer: curvature read: min=0.000000 max=0.000000" ps. rh.orig and rh.qsphere exist in surf directory!!!!

2 1

pial surface not plump
by Josh Snyder 23 Jun '04

23 Jun '04

Hi, The pial surface generated in one of my recons doesn't look standard. Part of it looks plump as usual, but mostly the anterior portion looks more like the smoothwm surface -- the gyri look too thin. I've checked in tkmedit, and it's hard to tell, but it does look as though in some areas it's choosing the wrong cut-off points between grey matter and empty space, leaving grey matter outside the pial surface. Does anyone know what might cause this, or have a recommendation for dealing with it? Thanks, Josh

1 0

mri_convert problem
by Scott.H.Johnson＠Dartmouth.EDU 22 Jun '04

22 Jun '04

Hello, We are tryying to convert some raw images from our GE LX scanner into cor format using "mri_convert" with plans to do inflation etc. We have tried a number of ways (see below) to no avail. The program tells us it is "reading images from 001/I.001," but never comes back..... If anyone has any suggestions, I would be very grateful. I am certain that we are overlooking something really obvious..... mri_convert 001/I.001 ./anat1 mri_convert -ot cor 001/I.001 ./anat1 note that we have the structure subjectID/mri/orig within the dir "orig" we have 001 002 003 and each contains a 124 slice spgr consecutively numbered starting with I.001 "orig" also contains empty anat1 anat2 anat3 dirs Many thanks, Scott

3 2

mritotal / Out of memory
by Patrice Seyed 22 Jun '04

22 Jun '04

Hi, I am experiencing the following error when running recon-all relating the mritotal: mritotal -verbose -debug -clobber -protocol icbm orig/orig.mnc transforms/talair ach.xfm Out of memory! ERROR: mritotal failed I found this brought up as an issue on the freesurfer mailing list but not a posted solution. I have been advised this is related to RH 9.0 and also that it is fixed in the recent snapshot... Here are my current specs: Feb. 18th, 2004 Snapshot/full install of freesurfer from (http://surfer.nmr.mgh.harvard.edu/download.html) Is there a snapshot/patch more recent than this for freesurfer? Caosity (http://www.caosity.org/),which is based off the current Red Hat EL 3.0. Re: the LANG variable in my current tcsh environment is: $ echo $LANG en_US.UTF-8 I edited the beginning of recon-all to reflect these variables as was suggested: set LANG = en_US.UTF-8 set LANGVAR = en_US.UTF-8 My next run seems to have crashed the system right around where swap space has run out (interesting): Mem: 513464 506716 6748 0 2040 7316 -/+ buffers/cache: 497360 16104 Swap: 1044176 1044044 132 total used free shared buffers cached Mem: 513464 505952 7512 0 2024 7180 -/+ buffers/cache: 496748 16716 Swap: 1044176 1044176 0 Thanks for any suggestions. Regards, Patrice Seyed Linux System Administrator - SIG RHCE, SCSA Boston University School of Medicine

5 4

Signal to Noise Values
by Chekema Nayasha Prince 22 Jun '04

22 Jun '04

Is there a way to get the SNR values from a region in the brain through tkmedit? or even a contrast to noise value? -Keyma

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer June 2004