Freesurfer October 2006

freesurfer@nmr.mgh.harvard.edu

41 participants
61 discussions

tksurfer problem
by Ran Tao 18 Oct '06

18 Oct '06

Hi, I have a problem about tksurfer command. After installing the freesurfer step by step according to instruction, I followed the tutorial to do some tests. However, when I tried to use “tksurfer good_output lh inflated” command on the webpage https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/OutputData, the result is not the image on the page, the image I got is attached to this email. At the same time, The tkmedit works fine. I have downloaded the tutorial data. Is there any other demo data should be downloaded? Anyone help will be greatly appreciated. Thanks, Ran ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ The error message is as follows. surfer: current subjects dir: /home/sci/rantao/myinstalls/freesurfer/subjects/buckner_data/tutorial_subjs surfer: not in "scripts" dir ==> using cwd for session root surfer: session root data dir ($session) set to: surfer: /home/sci/rantao Talairach transform /autofs/space/birn_045/users/BWH/buckner_data/group_study/103/mri/transforms/talairach.xfm does not exist ... Loading tal xfm file /home/sci/rantao/myinstalls/freesurfer/subjects/buckner_data/tutorial_subjs/good_output/mri/transforms/talairach.xfm surfer: Reading header info from /home/sci/rantao/myinstalls/freesurfer/subjects/buckner_data/tutorial_subjs/good_output/mri/T1.mgz surfer: vertices=125352, faces=250700 Talairach transform /autofs/space/birn_045/users/BWH/buckner_data/group_study/103/mri/transforms/talairach.xfm does not exist ... Loading tal xfm file /home/sci/rantao/myinstalls/freesurfer/subjects/buckner_data/tutorial_subjs/good_output/surf/../mri/transforms/talairach.xfm

2 1

System requirements
by Oddbjørn Sæther 18 Oct '06

18 Oct '06

Hi, concerning system requirements for installation of FreeSurfer: System requirements says: "A 2GHz or faster processor, at least 2GB of RAM, and a 3D graphics card with accelerated OpenGL drivers, is recommended. Freesurfer is highly CPU and memory intensive (and moderately disk intensive), so concentrate on boosting those performance aspects (more memory is better). Graphics requirements are minimal." We plan to install FreeSurfer on a Linux server (probably SuSe 9), and run from Windows PCs via FreeNX/NXclient. 1. Will that work? 2. What system requirements is advisable with 3-6 users? 3. Is 3D graphics cards necessary on the Windows PCs (or how does this work)? Thanks in advance! Oddbjørn Sæther Dept. of Neuroscience Norwegian University of Science and Technology

2 1

Rendering results of custom fMRI analyses on FreeSurfer surface
by raij＠nmr.mgh.harvard.edu 17 Oct '06

17 Oct '06

Dear FreeSurfers, We are running a FS-FAST group analysis where the individual subjects are morphed to an average brain via spherical surface. The output of that analysis (sig statistical maps, left hemisphere) is saved in $STUDY_DIR/<groupanalysisdir>/bold/<analysisname>/<smoothingdir>/<contrastname>/sig-lh_XXX.bfloat, Question 1: In the above, XXX is usually the slice number. However we have 25 slices in the input data, and our <contrastname> folder contains only bfloat numbers 000-006. I guess morphing the individual data to a sperical surface for group analysis changes the original slice structure, and hence the numbering. I guess we have seven "slices" because in the command func2sph-sess we defined -icoorder 7 (icosahedron order). Correct? Question 2: We use the above sig-lh_XXX.bfloat files as input to a custom matlab analysis. Remember we have XXX=000-006. What should the output of the matlab analysis look like so that rendering it onto a FreeSurfer surface will work? Is there any documentation related to this? Thanks! -Tommi --- Tommi Raij, M.D., Ph.D. Bldg 149, 13th St Charlestown, MA 02129 U.S.A. raij(a)nmr.mgh.harvard.edu

1 0

about drawing label with tksurfer
by Lei.Zhang＠uniklinik-freiburg.de 16 Oct '06

16 Oct '06

Hi, Dear FreeSurfers, >From the mailing list I got to know that the label can be drawn manually. I tried on the flatten occiptal surface with tksufer, the purpose of this is to draw the area of some functional activations for ROI analysis, -sesslabel. I loaded the flatten patch in tksufer and also the functional overlay, then around the area of the activations I make closed path -> custom fill ->save selected label. Now, I want to know: 1) Am I right to create the new label? 2) Can this label will be the structural constraint for ROI process? Thanks in advance, Lei

1 0

questions about func2roi-sess
by Lei.Zhang＠uniklinik-freiburg.de 16 Oct '06

16 Oct '06

Hi, Dear FreeSurfers, I want to do ROI analysis with func2roi-sess to process my Block Designed data, now I have some confusions with it. My block designed functional scan includes four different conditions and their correspnding null states. Now, for func2roi-sess options: -maskcontrast, i really don't know which contrast I should use, because in a single run I have 4 conditions for different tasks, and the contrasts were trial1_vs_null, trial2_vs_null, trial3_vs_null and trial4_vs_null, how to decide and select? -maskmap, sig, minsig, fsig. Which one I should use for my blocked data? And I tried maskcontrast with trial1_vs_null, and I got the .dat file which was created by roisummary-sess, in the file I found the 9th value was 4 conditions, it sounds right for my data? and the 10th value of estimates per condition was just the number of points in time window, but this value looks like an ER analysis, the right number for block design should be 1 according to the manual which in the directory of freesurfer/docs? Any hints would be appreciated! Thanks, Lei

1 0

why no answer?
by Joongnam Yang 16 Oct '06

16 Oct '06

I posted the following question again, since I didn't get any response. I am trying to overlay group-averaged data (FSL) on an averaged anatomical brain (Freesurfer). I did the following, mri_vol2surf --src AVG.gfeat/cope1.feat/thresh_zstat1.img --src_type analyze --srcreg register.dat --hemi lh --projfrac 0.5 --out AVG.gfeat/cope1.feat/thresh_zstat1-lh.w --out_type paint and got the following error message, MRISreadCurvature: could not open /home/nam/AVERAGE/average/surf/lh.thickness No such file or directory In the surf directory, I DID find "lh.avg_thickness". Would someone please let me know how to remedy this? Nam.

4 3

Re: [Freesurfer] Intensity/Conversion problem
by Derin Cobia 13 Oct '06

13 Oct '06

Bruce, I've been getting values from around 188-255 in the wm. -Derin

2 6

mri_convert on 64bit vs 32bit
by Niels Bergsland 12 Oct '06

12 Oct '06

Hi, I am experiencing quite a strange problem. I have a set of DICOMs that when I run the following on a 32bit machine, it works fine: mri_convert 031G0110.101.dcm 001.mgz When I run the exact same command (from within the same directory) on the 64 bit machine, I get the following error: "Currently non-Siemens mosaic image cannot be handled. Exit" (I should note that these are not actually mosaic images, they are individual) mri_convert --all-info on the 32 bit machine yields the following: mri_convert --all-info ProgramName: mri_convert ProgramArguments: --all-info ProgramVersion: $Name: stable3 $ TimeStamp: 06/10/12-20:37:40-GMT CVS: $Id: mri_convert.c,v 1.121 2006/02/22 05:39:36 greve Exp $ User: niels Machine: freya.bnac Platform: Linux PlatformVersion: 2.6.9-42.0.2.ELsmp CompilerName: GCC CompilerVersion: 30400 and on the 64 bit machine: mri_convert --all-info ProgramName: mri_convert ProgramArguments: --all-info ProgramVersion: $Name: stable3 $ TimeStamp: 06/10/12-20:38:29-GMT CVS: $Id: mri_convert.c,v 1.121 2006/02/22 05:39:36 greve Exp $ User: niels Machine: giza Platform: Linux PlatformVersion: 2.6.9-22.ELsmp CompilerName: GCC CompilerVersion: 30400 We have used the 64 bit install extensively before and never run into this issue. Any insight? Thanks! Regards, Niels Bergsland

2 2

mri_extract_label
by Behrang Nosrat Makouei 12 Oct '06

12 Oct '06

Hi everyone, My goal is to have the cortical segmentation in subject's native space in analyze format volume images. the series of commands i run are: mri_extract_label aparc+aseg.mgz LABELNUM outVolume.mgz mri_convert -rt nearest -rl rawavg.mgz outVolume.mgz outLabelSegmentation.img the problems I have are: 1- the segmentations I get are noisy 2- the segmentations are not as good as they look like in Surface segmentations and I can clearly see some holes and discontinuities at my final volumes. any suggestions? best, -- Behrang Nosrat Makouei, Engineering Science, MICA Lab (MIAL), TASC 9400, SFU. cell: 778-885-5016

1 0

func2roi-sess error?
by Lei.Zhang＠uniklinik-freiburg.de 12 Oct '06

12 Oct '06

Hi All, I tried func2roi-sess with my data, it ran quickly and gave one message at the end 'Unknow user: greve', but no other errer reports. I checked the sessid directory and found it created the -roisef directory, but no files in it. I think there must be something wrong, associated with user? Any hints would be appreciated! Thanks, Lei

2 2

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer October 2006