Freesurfer

freesurfer@nmr.mgh.harvard.edu

5 participants
30076 discussions

Is there any way to measure the volume of substantia nigra and red nucleus via FS?
by nellie 02 Nov '16

02 Nov '16

Hi, FreeSurfer experts, I would like to know is there any command to measure the volume of substantia nigra, red nucleus, and anyother nucleis in brainstem by using FS ? Thanks in advance, Nellie

2 1

Re: [Freesurfer] Segmentation depends on image file format
by LaMontagne, Pamela 01 Nov '16

01 Nov '16

We are preparing a large dataset for sharing. It has already been processed and edited through freesurfer with the dicom files. We would like to share the MR scans in Nifti format instead of dicom, however we are encountering this difference in results. We are using dcm2niix to convert. We would like to stick with this program for conversion and the BIDS package. Any suggestions on how we can use this nifti output and get similar freesurfer results? Thanks Pamela Douglas N Greve<http://www.mail-archive.com/search?l=freesurfer@nmr.mgh.harvard.edu&q=from:…> Thu, 28 Jan 2016 09:35:38 -0800<http://www.mail-archive.com/search?l=freesurfer@nmr.mgh.harvard.edu&q=date:…> I suspect that it has to do with the image geometry information. In nifit, this is stored in quaterion format. In FS, this is converted to direction cosines. This conversion is not reversible. So when you mri_convert the DICOMS to nii, mri_convert will get the direction cosines from the dicom file and convert them to quaternions. When you run recon-all using the nii file, the quaternions are converted back to direction cosines. But this does not yield exactly the same as what was in the dicom file. This conversion does not happen when you start out with mgz files. In version 6 I changed the code to try to reduce this effect, but I think it will always be there. On 01/28/2016 10:49 AM, Bruce Fischl wrote: > can you run mri_diff on the orig.mgz in the two runs and send us the > output? > On Thu, 28 Jan 2016, Parzer, Peter wrote: > >> mri_convert >> ________________________________________ >> Von: freesurfer-boun...(a)nmr.mgh.harvard.edu >> <freesurfer-boun...(a)nmr.mgh.harvard.edu> im Auftrag von dgw >> <dgwake...(a)gmail.com> >> Gesendet: Donnerstag, 28. Januar 2016 15:55 >> An: Freesurfer support list >> Betreff: Re: [Freesurfer] Segmentation depends on image file format >> >> What software did you use to generate the NIFTI files? Many of the >> converters do different things to the axes of the data, and this does >> affect results. >> >> hth >> d >> >> On Thu, Jan 28, 2016 at 8:49 AM, Parzer, Peter >> <peter.par...(a)med.uni-heidelberg.de> wrote: >>> Hi, >>> >>> we have brain images stored as DICOM and as NIFTI files. For some time we >>> used to import the NIFTI files to freesurfer for segmentation. I was >>> curious if it makes a difference if we would import the DICOM files >>> directly, and was surprised that it actually did. Here are the commands we >>> used: >>> Pamela LaMontagne, PhD Clinical Research Coordinator Washington University School of Medicine Department of Radiology Phone: (314) 362-3487 Voice: (314) 669-4551 ________________________________ The material in this message is private and may contain Protected Healthcare Information (PHI). If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail. ________________________________ The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

1 0

White Matter Edits
by Kirkish, Gina 01 Nov '16

01 Nov '16

Hello, I am a staff research associate for a group that specializes in MRI of Multiple Sclerosis. We use freesurfer's reconstruction approach, however the white matter masks are not very good for the MS disease population. We have combined the FreeSurfer algorithm with ANTS in order to get more precise white matter masks – which in theory should improve the surface volumes and aparc+aseg masks. When we have tried putting the ANTS white matter mask into freesurfer and re-running recon-all (recon-all -autorecon2–wm –autorecon3 –w <subID>), the wm_surface and aparc+aseg mask do not seem to follow the white matter surface volume (see attached pictures for an example). How do we insure that the white matter mask is taken into consideration? I know that edits to the white matter mask are done with a value of 255 but the actual white matter volume is 110 – could this be the problem? Let me know if I need to explain anything further. [cid:F06B2C58-32ED-45A4-9FB4-27D90ECFA496][cid:CC736D56-1123-458E-8ADF-D6372F0ACCD9] Thank you, Gina

1 0

Re: [Freesurfer] BBR not recommended for PETSurfer?
by Elijah Mak 01 Nov '16

01 Nov '16

Hi Christopher, We have very positive experiences with mri_coreg too. http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg47135.html Quite a bit of manual tweaking had to be done along with BBR for our AV and PK BPnd data. It works just fine on the PIB SUVRs though. Then, we switched to mri_coreg and manual tweaking was hardly necessary after that. It truly works wonders :) Best Wishes, Elijah

2 1

tkregister2 + mri_vol2vol do not transform
by Chiara Giacosa 01 Nov '16

01 Nov '16

Dear experts, I am new to freesurfer and I am struggling making transformation of images from one subject to another as well as to a template. I have tried different ways, but I always end up with the original image as if the transformation was never applied. I used tkregister2 from command line and mri_vol2vol. I noticed that the transformation matrices are always the identity matrix in the first 3 columns/rows, no matter what transformation I tell the program to do. I don't think that this makes sense so I wonder if I am doing something wrong or if there might be an issue with the version of tkregister I have. My *commands *are: tkregister2 --mov $subj/mri/orig.mgz --targ my_template/mri/orig.mgz --regheader --reg $subj/reg/orig2templ_tkreg.dat --noedit mri_vol2vol --mov $subj/mri/rh.hand_manual.nii.gz --targ my_template/mri/orig.mgz --o $subj/mri/rh.hand_subj_in_mytempl.mgz --reg $subj/reg/orig2templ_tkreg.dat *Output Matrix *$subj/reg/orig2templ_tkreg.da: subject-unknown 1.000000 1.000000 0.150000 1.000000000000000e+00 0.000000000000000e+00 0.000000000000000e+00 -1.370635986328125e+00 0.000000000000000e+00 1.000000000000000e+00 0.000000000000000e+00 -4.750839233398438e+00 0.000000000000000e+00 0.000000000000000e+00 1.000000000000000e+00 -1.000000000000000e+00 0 0 0 1 round Contents of my $FREESURFER_HOME/build-stamp.txt file: freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0 Any suggestion will be much appreciated! -- Chiara Giacosa Ph.D. candidate Concordia University

3 6

Re: [Freesurfer] Questions about outputs in Longitudinal Stream
by Tamara Tavares 01 Nov '16

01 Nov '16

Thank you for the clarification Martin I have a follow-up question. When I load the base to make sure it doesn't look blurry, should I check the norm.mgz or brainmask.mgz? Online says norm.mgz but I just want to make sure. Similarly, when I open the base and the longitudinal runs to see whether they align, again should I load the brainmask.mgz for these as well? Thank you in advance for your help, Tamara

1 0

Re: [Freesurfer] BBR not recommended for PETSurfer?
by Douglas Greve 01 Nov '16

01 Nov '16

Hi Christopher, yes, my position on BBR for PET has changed. I ran an analysis on FDG from an HR+ using BBR and coreg, and the results were not different enough to warrant choosing one over the other. However, coreg performed better on simulated FDG HR+ data, so I'm recommending coreg over BBR, at least for typical resolution scanners like the HR+. For high res (eg, HRRT), BBR may still be better. But for data with a PSF of around 6mm or higher, I would go with coreg. BBR remains the method of choice for all MRI volumes. doug On 11/1/16 10:21 AM, Schwarz, Christopher G., Ph.D. wrote: > > Dr. Greve, > > I noticed on https://surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer > <https://surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer> that you’re > recommending mri_coreg instead of bbregister for PET-MRI alignment in > PETSurfer. This seems inconsistent with your 2014 NeuroImage paper > http://www.sciencedirect.com/science/article/pii/S1053811913012251 > “BBR is very robust with respect to image artifacts and so an ideal > candidate for PET-MR registration” > > Has your position on BBR for PET-MRI changed? Was mri_coreg later > found to be more robust? > > Thank you for your time, > > Christopher Schwarz > > *--* > > *Christopher G. Schwarz, Ph.D.*| Research Associate, Diagnostic > Radiology | Opus Center for Advanced Imaging Research | 507-538-4867 | > schwarz.christopher(a)mayo.edu <mailto:schwarz.christopher@mayo.edu> > > *Mayo Clinic*| 200 First Street SW | Rochester, MN 55905 | > http://mayoclinic.org <http://mayoclinic.org/> >

1 0

Freesurfer recon-all error after installing
by Lehr, Albert 01 Nov '16

01 Nov '16

Dear FreeSurfers, I just recently installed FreeSurfer and as Saeed ( http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg49263.html) had troubles running recon-all. My info: 1) FreeSurfer version: freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0 2) Linux version: Linux unanderra 4.4.0-45-generic #66-Ubuntu SMP Wed Oct 19 14:12:37 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux I attach my recon-all.log but I guess it is very similar to Saeed's. Also running mri_nu_correct.mni --n 1 --proto-iters 1000 --distance 50 --no-rescale --i orig.mgz --o orig_nu.mgz gives the exact output and error code as he had. I already read that downloading the Development Version should get the job done, but noticed some stuff and wanted to ask about it: Running "recon-all -s bert -all" , the following lines are shown and the build stamps do not match. INFO: FreeSurfer build stamps do not match Subject Stamp: freesurfer-Linux-centos4-stable-v5.3.0beta-20130503 Current Stamp: freesurfer-Linux-centos6_x86_64-stable-pub-v5.3.0 INFO: SUBJECTS_DIR is /usr/local/bin/freesurfer/subjects Actual FREESURFER_HOME /usr/local/bin/freesurfer -rw-r--r-- 1 albert albert 513977 Oct 31 16:30 /usr/local/bin/freesurfer/subjects/bert/scripts/recon-all.log Linux unanderra 4.4.0-45-generic #66-Ubuntu SMP Wed Oct 19 14:12:37 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux INFO: current FREESURFER_HOME does not match that of previous processing. Current: /usr/local/bin/freesurfer Previous: /space/freesurfer/centos4.0/stable Equally in the attached recon-all.log file "FREESURFER_HOME /space/freesurfer/centos4.0/stable Actual FREESURFER_HOME /autofs/cluster/freesurfer/centos4.0/stable5" is written and although I downloaded the centos6-version and not the centos4.0 version. Might this be a problem or is the error solely due to the new perl versions? Just wanted to bring it to your attention and hear your feedback. I am currently downloading the 6.0.0-beta version, which hopefully makes recon-all work again. Have a great day, Albert

2 1

freesurfer preproc output
by Bernardoni, Fabio 01 Nov '16

01 Nov '16

Dear all, after running a freesurfer preprocessing, in each subject subdirectory, the following files are produced: lh.avg_curv lh.curv lh.curv.pial (and the analogue for the right hemisphere) what is the content of these files? sure, I can guess they represent the local curvature. but which one? the intrinsic? the extrinsic? what is the relation with the measures: gauscurv, meancurv, foldind, curvind obtained through the aparcstats2table? It was quite surprising to me that on the wiki no precise definition is present, although one can guess that gauscurv is the intrinsic and meancurv the extrinsic curvature. what are foldind and curvind by the way? how can one get the intrinsic and extrinsic curvatures from the lh.curv files? since these are all very useful measures, it would be of much help for me and other users (I believe) if some more intormation could be written on the wiki. thanks for any help, fabio Dr. Fabio Bernardoni wiss. Mitarbeiter Psychosoziale Medizin und Entwicklungsneurowissenschaften Tel. +49 (0)351 458-5245 Fax +49 (0)351 458-7206 URL http://www.uniklinikum-dresden.de/psm; www.transdenlab.de Universitätsklinikum Carl Gustav Carus & Medizinische Fakultät an der Technischen Universität Dresden Anstalt des öffentlichen Rechts des Freistaates Sachsen Fetscherstraße 74, 01307 Dresden http://www.uniklinikum-dresden.de Vorstand: Prof. Dr. med. D. M. Albrecht (Sprecher), Wilfried E. B. Winzer Vorsitzender des Aufsichtsrates: Prof. Dr. med. Peter C. Scriba USt.-IDNr.: DE 140 135 217, St.-Nr.: 203 145 03113

1 0

tksurfer viewing issues
by Paige Scalf 01 Nov '16

01 Nov '16

Hi Bruce, I'm analyzing phase encoded retinotopy, and it is not clear to me how to load complex overlays in freeview. Perhaps I am missing something? Best, Paige Hi Paige we have stopped development and most support of tksurfer. Can you use freeview instead or is there still some functionality that you need that is unique to tksurfer? cheers Bruce On Thu, 29 Sep 2016, Paige Scalf wrote: >* Hello, *> >* I too am having trouble with tksurfer displaying the surfaces upside down *>* and back to front, but believing that they are correctly oriented (i.e. *>* Talairach coordinates change in the wrong direction wrt to the surface but *>* in the correct direction wrt the display window). I"m also running a CentOs *>* flavor. Did anyone manage to solve this issue? *> >* All the best, *>* Paige* -- Paige E. Scalf, Ph.D. Lecturer Department of Psychology Durham University Psychology Department Durham University Science Laboratories South Road DURHAM DH1 3LE

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