Freesurfer

freesurfer@nmr.mgh.harvard.edu

11 participants
30080 discussions

Fixing WM intensity normalization via control points
by John Sheppard 24 Sep '08

24 Sep '08

Hello, I have some basic questions with regards to adding control points to fix the WM intensity normalization. I am running some subjects through autorecon 1 and 2 only (we are not concerned with subcortical structures). The reconstructed volumes had a lot of missing grey matter in the lateral temporal lobe, and my intention is to fix this by adding WM control points (the intensity is also off for WM in places). Long story short, I have the control points added and saved, and then I run the subjects through recon-all autorecon2-cp. 1) Am I correct that autorecon3 will not be necessary if I am only including cortical structures? (I see it is included in the command line on the control point tutorial.) 2) Once I run the subjects through autorecon2-cp, when I open them in tkmedit the segmentation still has not changed - the control points are still there as before. Do I need to run the subjects through some other processing step before I will see the changes? (I am doing tkmdedit SUBJECT brainmask.mgz lh.white -aux T1.mgz -aux-surface rh.white) 3) Based on the workflow in http://surfer.nmr.mgh.harvard.edu/fswiki/RecommendedReconstruction, it seems that control points need to be added directly after autorecon1. I am actually adding the CPs on subjects that have gone through autorecon2 already. Is this a problem or is it still possibe? (If a problem I can reprocess the data.) Sorry if these are painfully obvious questions, but I seem to be slightly confused with the processing pipeline. Much thanks, John Sheppard

1 0

Re: [Freesurfer] error: mris compute white volume
by Hans Johnson 24 Sep '08

24 Sep '08

Nick, I am tryng to get the updated executables for mri_segstats and mris_wm_volume that are described in your mail message: http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg08009.html ==================================================== Christian, Updated mri_segstats and mris_wm_volume executables for centos 32/64bit can be downloaded from here: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/fsdev/nicks/stable Nick ==================================================== 1) Will you please point me to the correct version of these files? I'm currently attempting to run about 5000 freesurfer jobs on the teragrid using the BIRN "Grid Wizard Enterprise" batch processing tool, and these 6Gigabyte log files are killing my disk quota requiring a lot of extra manual cleanup. 2) I'm using a series of platforms to get the job done, so I would really appreciate binaries for the following 3 platforms: centos4, centos4_x86_64, macintel-leopard. Separage Issue: The recon-all "-nolog" option does not propogate properly to the /opt/freesurfer/bin/recon-all.makefile. The problem is that recon-all internally uses "make -f $FREESURFER_HOME/bin/recon-all.makefile ${make_flags}" which then calls the recon-all again. The problem is that the nested call of recon-all from the makefile does not get the command line flags for redirecting the log files. You help is always appreciated. Thanks, Hans -- Hans J. Johnson, Ph.D. Hans-johnson(a)uiowa.edu 278 GH The University of Iowa Iowa City, IA 52241 (319) 353 8587

1 0

Count non-zero voxels without matlab
by Jeff Sadino 24 Sep '08

24 Sep '08

Hello surfers, We are trying to normalize volumes of different regions of the brain to brain size. By brain size, we would be shooting for everything inside the skull - brain, CSF, and dura. After looking at the four different measurements in aseg.stats and comparing what each of these measures include and also how much each of these measures vary between scans on the same subject, none of them really achieve the consistency that we would like. ICV had a variability of 7.4% with 5 subjects each with 3 scans. The others were at about 3%. We are thinking that if we count the non-zero voxels in brain.mgz, this should be the most accurate measure of ICV. It will include a little bit of the skull and maybe some outside the skull, but since the size of the skull itself should be constant from scan to scan, that should not really matter. This measure gave a variability of only 1.51% Does anybody see any potential errors in using this approach to measure ICV, or suggest improvements or verify its validity? Thank you so much! Jeff Sadino mri_label_volume -a brain.mgz should do the trick. cheers, Bruce On Mon, 22 Sep 2008, Jeff Sadino wrote: > > Hello surfers, > > I would like to count the number of non-zero voxels in the image brain.mgz. Does anybody have any ideas what the easiest way to do this would be without using matlab? Thank you all! > > Jeff Sadino _________________________________________________________________ See how Windows connects the people, information, and fun that are part of your life. http://clk.atdmt.com/MRT/go/msnnkwxp1020093175mrt/direct/01/

2 1

reg-feat2anat misregistration (potentially resolved?)
by Fred Dick 23 Sep '08

23 Sep '08

Dear Doug et al., a while back there was a post on peculiar registration problems using reg-feat2anat, e.g., the registration using reg-feat2anat was quite off (seriously rotated or similar), whereas barebones flirt on the same volumes did better. I didn't spot a resolution on the message boards, but I have found that the EPIs end up being rotated incorrectly (around A-P axis) when I run flirt with the initial registration (from tkregister2-cmdl), but not with the same command minus -init. I didn't see any hidden option flags in reg-feat2anat for blowing off -init (although did spot one in fslregister), so am guessing that other people haven't found this to be a problem. I should mention that I'm using reg-feat2anat on a partial volume scan - there may not be the same issue with whole-head EPIs... Apologies if I missed a previously posted solution, best, Fred

1 0

white.surf doesn't follow wm.mgz
by dahlia＠nmr.mgh.harvard.edu 23 Sep '08

23 Sep '08

Hi, What should I do when the white volume looks right (e.g. filled in where needed) but the white surface doesn't follow the boundary of the volume? I've done recon-all with autorecon2-wm and autorecon3. Thanks! Dahlia.

2 5

Count non-zero voxels without matlab
by Jeff Sadino 22 Sep '08

22 Sep '08

Hello surfers, I would like to count the number of non-zero voxels in the image brain.mgz. Does anybody have any ideas what the easiest way to do this would be without using matlab? Thank you all! Jeff Sadino _________________________________________________________________ Want to do more with Windows Live? Learn “10 hidden secrets” from Jamie. http://windowslive.com/connect/post/jamiethomson.spaces.live.com-Blog-cns!5…

2 1

Spherical visualization
by Pedro Paulo de Magalhães Oliveira Junior 22 Sep '08

22 Sep '08

Sorry for the question but in order to generate the visualization of the spherical mapping of, for instance, cortical thickness we should use which command line in TKSurfer? -- ----------------------------------------------------------- Pedro Paulo de M. Oliveira Junior Diretor de Operações Netfilter & SpeedComm Telecom

2 2

fsaverage across versions
by Steinmann, Iris 22 Sep '08

22 Sep '08

Hi, we have several reconstructed brains, which were processed by freesurfer 3.0.5. We also kept using the fsaverage data from 3.0.5 in order to be consistent. We wish to do fMRI analysis with freesurfer/fs-fast 4.0.5 and found that talairach coordinates differ slightly when using the fsaverage from version 3.0.5 and the current fsaverage from 4.0.5. Yet, the both volumes appear pretty similar, the newer one maybe a little bit smoother. We would like to know what constitutes the exact difference between these two fsaverage datasets and whether it is possible to use the new version for analysis with the "old" fsaverage data without getting inconsistent results. Thanks a lot, Iris

2 1

Methodology Papers.
by Matt Clarkson 22 Sep '08

22 Sep '08

Dear everyone, Hopefully someone can help me. If I do "recon-all --help" i can see a list of 31 steps for the main cortical thickness pipeline. Does anyone have a list that links each of the 31 steps to the most current methodology paper? I've been reading most of the methodology papers referenced in the wiki, but the software has clearly evolved over the years, and even though you can read the wiki, and trawl through log files, I want to be absolutely sure that for each step, I know exactly what algorithm is applied. Has anyone gone through this before? Thanks in advance, Matt Matt Clarkson Senior Research Associate Dementia Research Centre UCL Institute Of Neurology Queen Square London WC1N 3BG m.clarkson(a)ucl.ac.uk Tel: 08451 555 000 ext. 723653 Fax: 020 7676 2066

4 6

Re: [Freesurfer] automating tkmedit
by Graham Wideman 20 Sep '08

20 Sep '08

... also if anyone might find these helpful, there's an article that may be of interested here: http://surfer.nmr.mgh.harvard.edu/fswiki/QaImageScripts This shows tcl scripts to do something substantial in tkmedit and tksurfer, and bash scripts to activate FreeSurfer and the tcl scripts. Hope that helps, Graham

1 0

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