Freesurfer

freesurfer@nmr.mgh.harvard.edu

1 participants
30089 discussions

Control point edits adversely affect cortical surfaces using FS 5.1
by Vanessa Louise Cropley 11 Oct '12

11 Oct '12

Hi all, I have seen others post a similar problem on the freesurfer mailing list but I could not find a solution. When I edit a volume using control points, I am noticing that the use of CP edits adversely effects the wm/gm surfaces in the resulting re-processed image. Like others have reported, the surfaces are adversely affected in locations distant from the CP edits. The temporal lobes are particularly affected. We are using freesurfer 5.1. I have noticed the following things: 1. The problem occurs irrespective of the number of control points added, i.e. whether >100 or <10 control points added 2. The problem occurred when only one CP was added in a voxel that was clearly wm 3. The problem did not occur after reprocessing with FS 4.5 4. I have tried the flag to disable mri_ca_normalize when reprocessing with –autorecon2. This did not seem to help. I used the command: recon-all –nocanorm –autorecon2-cp –subjid $SUBJ I have screenshots available that illustrate the changes in the surfaces after reprocessing. Could anyone help with this problem? Thank-you, Vanessa

2 1

Boundary problem
by youngmin huh 11 Oct '12

11 Oct '12

Dear all, I'm working with some 3T images and got some bad segmentation problem. I ran recon1 with nuintensitycor-3T, recon2, and recon2-cp after adding some control points. I attached an example. As you can see, it has kind of scrambled white matter boundaries. I wonder how I could fix this. Thanks, youngmin Huh

2 1

Re: [Freesurfer] cortical parcellation question
by Bruce Fischl 11 Oct '12

11 Oct '12

Hi Daniel, no, sorry. The gca is not a cortical atlas. Those labels are stored in the .gcas file and aren't in the same space (in fact, they are in a surface-based spherical coordinate system and not in the volume at all, so not easily combinable). Bruce On Thu, 11 Oct 2012, Daniel Ferreira wrote: > Thanks for the detailed information. > > The mean intensity image (mean_intensity.nii) generated from frame #1 is > almost exactly what I need. This image has all the sub-cortical regions > assigned a discrete integer number (i.e. all the voxels in the left > cerebelllar grey matter = 8 and right = 47). However the whole of the > neocortex is defined as left (intensity = 3) and right (intensity = 42) > only, there are no cortical sub-divisions. > > Is it possible to have the individual neocortical regions assigned a > discrete integer number; (obviously not one already assigned to the other > regions in the mean_intensity.nii image)? For example all inferior temporal > voxels = 9; middle temporal voxels = 10; superior temporal voxels = 11. So > each of the regions defined in the "colortable_desikan_killiany.txt" file > found in the FREESURFER_HOME/average/ directory are also defined in the > mean_intensity.nii image? > > Thanks > > > 2012/10/6 Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> > Hi Daniel > > it's in .gca format, and includes information that wouldn't fit > in a nifti file (e.g. markov relationships). You can use > mri_convert to extract some of the basic info though (e.g. the > most likely class at each location and the mean intensity of the > most likely class) using mri_convert. For example: > > mri_convert $FREESURFER_HOME/average/RB_all_2008-03-26.gca#0 > ./labels.nii > mri_convert $FREESURFER_HOME/average/RB_all_2008-03-26.gca#1 > ./mean_intensity.nii > > > where the #<n> syntax mean extract the nth frame from the 4D gca > volume. Frame 0 is the most likely label and frame 1 is the mean > intensity. Note however that there is lots more info in the .gca > that this doesn't capture such as the MRF parameters I mentioned > above, class variances, other class probabilities, etc.... > cheers > Bruce > > > > > On Thu, 4 Oct 2012, Daniel Ferreira wrote: > > Thanks Bruce, > When you say the atlas itself is included in the > Freesurfer distribution, do > you mean there is a 3D version of it available in > the download? > Also, I am not looking for the individual subjects' > images, I just want a > copy of the final atlas as a 3D image in > analyze/nifti format. (i.e. in the > same format as the Harvard-Oxford atlas that is > downloaded with FSL). > > Thanks a lot. > > Daniel Ferreira > > > > 2012/10/3 Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> > Hi Daniel, > > 1. Do you mean the atlas or the parcellation > of your subjects? > We cannot distribute the individual subjects > in the atlas as the > data was collected under an IRB long ago that > didn't include > this. The atlas itself (summary statistics > derived from those > subjects) is included in the FreeSurfer > distribution. > > 2. The subcortical areas are the same, but not > the cortical > ones. That cortical parcellation is quite > coarse, and while we > did implement it we were never given > permission to distribute > the automated version, so it is not part of > FreeSurfer. The > boundaries that we define are thus different > than that one (and > since the Harvard-Oxford one is based on > volumetric labeling of > folding patterns we found that it contained a > bunch of errors in > which e.g. the wrong side of a sulcus was > labeled) > > cheers > Bruce > > > On Wed, 3 Oct 2012, Daniel Ferreira wrote: > > Dear experts, > I have a couple of questions regarding > the cortical > parcellation atlas that > is used in Freesurfer: > > 1) I would like to know if it is > possible to obtain > the Desikan-Killiany > atlas in 3D analyze/nifti format. With > each region > of the brain assigned a > discrete integer number (for example the > all of the > voxels of the left > inferior temporal lobe have an intensity > = 1, or > the whole of the left > hippocampus = 2). > > 2) Are the Harvard-Oxford cortical and > subcortical > atlases that can be > downloaded with FSL the same as the > cortical > parcellation atlas that is used > in Freesurfer? (i.e. are the > neuroanatomical > boundaries the same; does the > left inferior temporal cortex in the > Freesurfer > cortical parcellation = the > left inferior temporal cortex in the > Harvard-Oxford > atlas downloaded with > FSL) > > Many thanks > > Daniel Ferreira > > > > > > > The information in this e-mail is intended only for > the person to whom > it is > addressed. If you believe this e-mail was sent to > you in error and the > e-mail > contains patient information, please contact the > Partners Compliance > HelpLine at > http://www.partners.org/complianceline . If the > e-mail was sent to you > in error > but does not contain patient information, please > contact the sender > and properly > dispose of the e-mail. > > > > > >

1 0

WM/GM boundary changes and control points
by EK Kim 10 Oct '12

10 Oct '12

I already uploaded this problem 2 days ago, reprocessing the subject to make sure about control points. However when I added the control points to the left hemisphere, the right hemisphere surface was changed severely. I used the command line, recon-all -autorecon2-cp -s subjectID. Could you tell me why does it occur and how to deal with it? Best, EK

2 3

assembling data
by Gregory Kirk 10 Oct '12

10 Oct '12

a have a bunch of data stored in the subj_id directories of the subjects for convieniance but it is already in the average surface space, but i want to assemble it into one file so i can run the glm. mris_preproc --fsgd subset.fsgd --target subset_average --hemi lh --meas winkler.mgh --out lh.surf.mgh does not work obviously as the subjects # verticies does not agree ERROR: number of vertices in /study5/aa-scratch/UTAH/freesurfer_subjects/AD33112_2/surf/lh.winkler.mgh does not match surface (163842,149346) because it thinks the surf is on the subj_id surface. how can i accomplish ? cheers greg

2 1

volumes per lobe
by Carolina Valencia 10 Oct '12

10 Oct '12

Hi FS users and developers, I need to obtain the volume per lobe (occipital, parietal, temporal, frontal) in a particular subjects and if it is possible divided by gray and white matter. I found similar topics in the mailing list, but I don't succeed to get this values. I follow the instructions for the FSv5.1 Using mri_annotation2label and mris_anatomical_stats but I got values for each hemisphere instead of lobes. mri_annotation2label --subject vol17 --hemi rh --lobesStrict rhlobesfile mri_annotation2label --subject vol17 --hemi lh --lobesStrict lhlobesfile then: mris_anatomical_stats vol17 rh mris_anatomical_stats vol17 lh I got values number of vertices = 115348 total surface area = 76188 mm^2 total gray matter volume = 282811 mm^3 average cortical thickness = 3.071 mm +- 1.034 mm average integrated rectified mean curvature = 0.149 average integrated rectified Gaussian curvature = 0.106 folding index = 4797 intrinsic curvature index = 391.1 Can somebody guide me? Thank you in advance, -- -- *Carolina Valencia * The information contained in this message and its attachments is confidential and intended for exclusive use of the recipients. If you received this email by error please notify the sender and delete it. The use, retention or distribution for any purpose is prohibited. Link is not responsible for damage caused by the contents of this message. The recipient should check for any presence of viruses or other defects.

2 3

(Ir?)Regular durations for recon-all -all
by Denis-Alexander Engemann 10 Oct '12

10 Oct '12

Dear Freesurfers, presumably this is quite a newbie's question but I was wondering whether my recon-all times can be plausible if it takes less then 10 or 12 hours to perform the entire routine on a local machine. I'm using freesurfer 64 bit on Debian testing equipped with 16GB of RAM and Intel Core i7 CPUs (3.4Ghz), build-stamp: freesurfer-Linux-centos4_x86_64-stable-pub-v5.1.0. On superficial inspection, the results seem ok but I fear I'm overlooking something important. Any pointers would be of great help. Cheers, Denis ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Juelich GmbH 52425 Juelich Sitz der Gesellschaft: Juelich Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Dr. Karl Eugen Huthmacher Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender), Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt, Prof. Dr. Sebastian M. Schmidt ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Kennen Sie schon unsere app? http://www.fz-juelich.de/app

1 0

Freesurfer and FSL
by Ana Arruda 10 Oct '12

10 Oct '12

Dear experts, I'm an MSc student and part of my final project is related to the measurement of the volume of the hippocampus. The last thing I have to do is to compare two softwares, FSL and Freesurfer. On Freesurfer, I've used the following commands: recon-all -i where_my_subject_is -autorecon1 -s output_name then, related to the same subject recon-all -autorecon2 -s output_name After approximately 12 hours, and after comparing the results of Freesurfer and FSL, I found that those obtained with Freesurfer are bigger than those obtained with FSL. According to the literature the results are the reverse. Do you know any idea what I'm doing wrong? Thank you for your attention, Ana Arruda

6 25

Dimension Issues with mri_surf2surf
by Aimi Watanabe 10 Oct '12

10 Oct '12

Hello, My input to freesurfer is: mri_surf2surf --cortex --fwhm-src 5 --noreshape --hemi rh --tval [$output file] --sval [$input file] --srcsubject [$subject id] --trgsubject fsaverage --tfmt nii When I reopen the resulting image with Nibabel, the dimensions are 27307 x 1 x 6 x 296. While 296 is the time axis, I was hoping to get 27307*6 = 163842 vertices. I have replaced '--noreshape' with '--reshape', '--reshape_factor=6' and the bash output looks proper: 121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Saving target data 121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Reshaping 6 (nvertices = 163842) 121010-16:49:28,430 interface INFO: stdout 2012-10-10T16:49:28.429912:Saving to [$output file] However, it does not change the resulting image. I have also tried other file types that are not 'nii', like 'mgh', but this does not help either. Please advise? Also, I could easily make a work-around, and reshape the image from a(27307, 1, 6, 296) to b(163842, 1, 296) in FORTRAN (column-major) order. I.E.: 0~27306 then add the next set 27307~54613. If I do this, would the original vertex map be preserved? Thank you!

2 1

hippocampal segmentation
by Hugh Pemberton 10 Oct '12

10 Oct '12

Dear freesurfer crew, I'm having a few issues with segmentation around the hippocampus, I have included a few screenshots of main and aux vol where the hippocampus has been excluded even with the addition of control points. Have you got any tips for rectifying this? It's happened a fair few times with my subjects. Thanks very much in advance and all the best, Hugh

2 1

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