Freesurfer

freesurfer@nmr.mgh.harvard.edu

11 participants
30080 discussions

Bug with aparcstats2table on aparc.a2005s parcellation - addendum
by Laluz, Victor 23 Oct '08

23 Oct '08

The body of my email was lost because of the attachments: here it is again for those who are curious. Sorry about that, I didn't know a text file attachment would do that. Not sure if this bug has been mentioned before, but there is a bug with using aparcstats2table for the aparc.a2005s parcellation map. If some of the subjects in the group have missing values of a certain type, aparcstats2table fails to designate a 0 for them. So for instance, it's often the case that a subject has no lh_unknown_volume in their aparc.a2005s.stats. Certain other, seemingly random areas will fail to print out as well. This will lead to shifts in the dataset, so that every volume after the missing value for a subject will be incorrect, and will need to be shifted for the blank. The only way I have found to reliably check that the data is valid is to generate seperate stats tables for each subject. Anyway, this is a serious problem for people who collect and analyse data with aparcstats2table using the aparc.a2005s parcellation map. Attached are two examples illustrating the problem. To make it clearer, paste the contents of both into excel, and look at the column mismatch; some of the subjects' data don't make it all the way to the end, and a shift in the values is clear. Victor Laluz : Imaging Coordinator : UCSF Memory and Aging Center 415-476-1680 : 415-476-1816 (fax)

2 1

convert multiple Analyze to single 4D nifti
by Xu Cui 23 Oct '08

23 Oct '08

Does anybody know how to convert a series of Analyze files into a single nii file? Each analyze file is a single 3D volume at a time point; the nii file would be 4d. Thanks, Xu -- ---------------------------------------------------------------- Xu Cui Department of Psychiatry and Behavioral Sciences Stanford University ----------------------------------------------------------------

1 0

Bug with aparcstats2table on aparc.a2005s parcellation
by Laluz, Victor 23 Oct '08

23 Oct '08

Hello, Not sure if this bug has been mentioned before, but there is a bug with using aparcstats2table for the aparc.a2005s parcellation map. If some of the subjects in the group have missing values of a certain type, aparcstats2table fails to designate a 0 for them. So for instance, it's often the case that a subject has no lh_unknown_volume in their aparc.a2005s.stats. Certain other, seemingly random areas will fail to print out as well. This will lead to shifts in the dataset, so that every volume after the missing value for a subject will be incorrect, and will need to be shifted for the blank. The only way I have found to reliably check that the data is valid is to generate seperate stats tables for each subject. Anyway, this is a serious problem for people who collect and analyse data with aparcstats2table using the aparc.a2005s parcellation map. Attached are two examples illustrating the problem. To make it clearer, paste the contents of both into excel, and look at the column mismatch; some of the subjects' data don't make it all the way to the end, and a shift in the values is clear. Victor Laluz : Imaging Coordinator : UCSF Memory and Aging Center 415-476-1680 : 415-476-1816 (fax)

1 0

creating "intermediate" surfaces
by Rudolph Pienaar 23 Oct '08

23 Oct '08

Hi all - Is there a way to create "intermediate" surfaces in FreeSurfer? By intermediate I mean "some fraction" between gray/white and pial. Cheers -=R -- Rudolph Pienaar, M.Eng, D.Eng / email: rudolph(a)nmr.mgh.harvard.edu MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging 149 (2301) 13th Street, Charlestown, MA 02129 USA

2 1

mri_convert
by sameer kumar 23 Oct '08

23 Oct '08

Hi all, I am new to FS and I would like to know the usage of mri_convert to convert the results of FreeSurfer into the analyze format. Thanks, Regards, Sameer Kumar Gottipati

3 7

FreeSurferColorLUT & aparc
by Juranek, Jenifer 23 Oct '08

23 Oct '08

Hi, Running fsv405 on RHEL5. I've encountered some unexpected numerical values wrt FreeSurferColorLUT.txt. Are my expectations off? 1) In the aparc+aseg.mgz file, intensity values for the cortical labels range from -32768 to 654, with negative values falling in some regions in the left hemisphere. 2) Identical results are obtained if mri_aparc2aseg is used to generate an output file: mri_aparc2aseg --s mysubject --o ./aparc2aseg.mgz 3) Values of aseg.mgz map perfectly well to FreeSurferColorLUT.txt Example values obtained from aparc+aseg.mgz are listed below: lh.superiortemporal=484 (expecting 1030 from FreeSurferColorLUT.txt) rh.superiortemporal=32767 (expecting 2030) Although these numerical values do not match the LUT, they are consistent across subjects. Many Thanks for any insights... Jenifer

2 5

questions about longitudinal analysis
by Wang, Xin 23 Oct '08

23 Oct '08

Hi, groups, I noticed the updates of longitudinal analysis in the new version of Freesurfer. However, I still have a couple of questions about the goal of longitudinal analysis. 1. The goal of longitudinal analysis is to detect the atrophy (thinning), but the transformation in the registration step will minimize any atrophy. An example of outcomes of FLIRT and FNIRT on FSL website exactly demonstrates that the atrophy disappears in the 'best' transformation (FNIRT) http://www.fmrib.ox.ac.uk/fsl/fnirt/index.html. In our experience, the number of vertex of post-transformed tp2 is very close to tp1. It is not clear to me how the difference in the original thickness between tp2 and tp1 is preserved in this transformation. One may argue that processing tp1 and tp2 independently does not solve this problem because the registration is also used in the routine recon. Is it true? 2. The new longitudinal process is still using volume-based registration if I read it correctly. Why not switch to surface-based since people have argued the advantage of surface-based over volume-based? Thank you for any clarification, Wang

2 1

Re: [Freesurfer] Error with mri_convert failing to write orig/001.mgz
by John Sheppard 23 Oct '08

23 Oct '08

Hey Krish, Thanks for your input. Good idea too, never thought about doing that. Tried this in command line and worked without a hitch. Then i realized if you just use command line what I did before works as well. The error must somehow be resulting from issues with tcsh or something. But I found a way to get it to work in tcsh shell script, going off of what you suggested. Here is what works for me in tcsh script: cd $FSsubjDir mkdir $subjID cd $subjID mkdir mri cd mri mkdir orig cd orig mv ../../Volume_ns+orig* ./ mri_convert --in_type brik Volume_ns+orig.BRIK 001.mgz etc etc That's probably a primitive way to write the script, but hey, it works. Thanks for the help! John On Wed, Oct 22, 2008 at 10:34 PM, Krish Subramaniam < krish(a)nmr.mgh.harvard.edu> wrote: > John > > Is orig/ already there ? Otherwise try > > mkdir orig > mri_convert --in_type brik Volume_ns+orig.BRIK 001.mgz > mv 001.mgz orig/ > > Thanks > Krish > > > > On Oct 22, 2008, at 10:01 PM, John Sheppard wrote: > > Hey everyone, >> >> I'm having some issues trying to run the volume pipeline using .BRIK input >> from afni (hence running mri_convert command directly). The problem occurs >> at the end of mri_convert at the very beginning when Freesurfer fails to >> write orig/001.mgz, and I'm not sure if I'm making a mistake without >> realizing it or if something weirder is going on. I've tried this on both a >> Mac and a linux station running RHEL, with the same results. I ran the >> script using sudo in RHEL so I do not think this is a permissions issue. >> >> The error occurs right when it tries to write orig/001.mgz: writing to >> orig/001.mgz... >> sh: orig/001.mgz: No such file or directory >> Broken pipe >> >> I'm not really familiar with working with pipes but I'm guessing I >> shouldn't have to deal with that. >> >> Just curious if anyone has encoutered this before and knows what's going >> on. Didn't find this issue in prior mailing list emails, so I hope this >> isn't a repeat question. >> >> Thanks! >> >> John Sheppard >> >> Recon output below: >> >> mri_convert --in_type brik Volume_ns+orig.BRIK orig/001.mgz >> $Id: mri_convert.c,v 1.146.2.3 2008/08/11 22:18:58 nicks Exp $ >> reading from Volume_ns+orig.BRIK... >> AFNI Header Information ============================================ >> DATASET_RANK : spatial dims 3, sub-bricks 1 >> DATASET_DIMENSIONS: (256, 256, 160) >> TYPESTRING : 3DIM_HEAD_ANAT >> SCENE_DATA : view type 0, func type 0, verify 0 >> ORIGIN : (-172.926697, 84.177513, -82.874474) >> DELTA : (1.000000, -1.000000, 1.000000) >> IDCODE_STRING : XYZ_FSWwmSca-raNtz0oRbtKow >> BYTEORDER_STRING : LSB_FIRST >> BRICK_STATS : min 0.000000 max 950.000000 >> BRICK_TYPES : short >> BRICK_FLOAT_FACS : 0.000000 >> ==================================================================== >> BRICK_STATS min = 0.000000 <--> actual min = 0.000000 >> BRICK_STATS max = 950.000000 <--> actual max = 950.000000 >> TR=0.00, TE=0.00, TI=0.00, flip angle=0.00 >> i_ras = (0, -1, 0) >> j_ras = (0, 0, -1) >> k_ras = (-1, 0, 0) >> writing to orig/001.mgz... >> sh: orig/001.mgz: No such file or directory >> Broken pipe >> Subject Stamp: freesurfer-Linux-centos4-stable-pub-v4.1.0 >> Current Stamp: freesurfer-Linux-centos4-stable-pub-v4.1.0 >> INFO: SUBJECTS_DIR is /mnt/win/MTLD1/autorecon >> Actual FREESURFER_HOME /usr/local/freesurfer >> Linux SLR10.soc.northwestern.edu 2.6.18-92.1.13.el5 #1 SMP Thu Sep 4 >> 03:51:01 EDT 2008 i686 i686 i386 GNU/Linux >> #-------------------------------------------- >> #@# MotionCor Wed Oct 22 20:45:32 CDT 2008 >> ERROR: no run data found in >> /mnt/win/MTLD1/autorecon/E-05-010-2-afni-test/mri. Make sure to >> have a volume called 001.mgz in >> /mnt/win/MTLD1/autorecon/E-05-010-2-afni-test/mri/orig. >> >> >> -- >> John Sheppard >> Undergraduate Student (Class of 2010) >> Department of Biomedical Engineering >> Northwestern University >> >> Phone: (630) 957-8321 >> E-mail: john-sheppard(a)northwestern.edu >> Address: 1927 Orrington Ave. Rm 8203 >> Evanston, IL 60201 >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer(a)nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> > > -- John Sheppard Undergraduate Student (Class of 2010) Department of Biomedical Engineering Northwestern University Phone: (630) 957-8321 E-mail: john-sheppard(a)northwestern.edu Address: 1927 Orrington Ave. Rm 8203 Evanston, IL 60201

1 1

mesh size question
by Tommaso De Marco 23 Oct '08

23 Oct '08

Hi all, I'm using the mri_watershed instruction to extract the skull and skin mesh, because I'm using them in a boundary element method, but the dimensions of these meshes are too big for my application (20480 faces and 10242 vertices for each ones). Do you know if there is a way to define a maximum mesh size for the mri_watershed instruction (or at least an equivalent instruction in which is possible to do it) ? I need of very coarse meshes ( 1000 faces for each ones for example). Thanks!

2 2

Rat template creation: MRIstripskull failed
by Mark Bouts 23 Oct '08

23 Oct '08

Hi, I'm trying to set up a rat brain template for analysis using the freesurfer tool. Before setting up the complete template I have to process a representable rat dataset with the available human template. With some much appreciated help I managed to get to the skull strip procedure . But the application ceases its operations at this point with the message: "White matter intensity 0 is lower than CSF intensity 21. Please examine input images. Will terminate ..." Evaluation using tkmedit does not show something a proper segmentation of the rat brain, however it does show a proper outline of the brain. Since a rat brain and a human brain are very dissimilar I'm not surprised by the fact that the segmentation is incorrect. However the brain is outlined ok. It is also known that the WM intensity values in the rat dataset is somewhat reduced to the GM structure. Which is not expected by the freesurfer tool, which expects WM to be highest intensity value. My question is: is there a way of circumventing this problem. For instance by defining the mean intensity values of each tissue class or.. Hope someone can help me? Thanks! Mark

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