Freesurfer

freesurfer@nmr.mgh.harvard.edu

30132 discussions

Talairach transformation
by Maria Felber 13 Sep '11

13 Sep '11

Hi Bruce, so we are back at the beginning. After I ran recon-all -all, I did exactly that and checked the xform using tkregister2. And here I always see that my talairach volume (the transform) is always smaller, which can be very well seen as the orig surfaces of the subjects are always a little outside the brain and reach well into the skull areas. Of course I can make the adjustments as descriped in the short instructions. But this also means I ran the recon-all precess anew for another 30 h per subject. And that is the reason, why I asked if this is normal that the tailarach transform is a) always smaller than the original brain, and b) if I can make any adjustments to prevent that from happening and thus saving another 30h per subject. In short: How acurate hat the talairach.xfm has to be and is it possible to change some parameters to make this transformation a little more accurate while running the recon-all process only once? Best, Maria ----- Original Message ----- From: "Bruce Fischl" <fischl(a)nmr.mgh.harvard.edu> To: "Maria Felber" <felber(a)cbs.mpg.de> Cc: freesurfer(a)nmr.mgh.harvard.edu Sent: Tuesday, September 13, 2011 4:07:55 PM Subject: Re: [Freesurfer] Talairach transformation Hi Maria you can use tkregister2 to check the accuracy of the talairach.xfm transform ("xform" for short), and correct it if you want. cheers Bruce On Tue, 13 Sep 2011, Maria Felber wrote: > Hi Bruce, > > thanks for the first response. I tried to find some information on the webpage about xform, but failed. So what is xform standing for and where/how can I check it? > As in the later processing stages I want to do localization and DCM-Modeling with my data, I guess the correct talairach coords are kind of important for later comparisons. > Thanks again, > Maria > > ----- Original Message ----- > From: "Bruce Fischl" <fischl(a)nmr.mgh.harvard.edu> > To: "Maria Felber" <felber(a)cbs.mpg.de> > Cc: freesurfer(a)nmr.mgh.harvard.edu > Sent: Friday, September 9, 2011 2:13:05 PM > Subject: Re: [Freesurfer] Talairach transformation > > Hi Maria > > as long as the tal xform is reasonable I wouldn't worry about it, unless > for some reason you care a lot about the accuracy of the talairach coords. > > cheers > Bruce > > p.s. not sure why it would be smaller > > On Fri, 9 Sep 2011, Maria Felber wrote: > >> Dear all, >> >> sorry, if that question came up before in the mailing list and I simple have been unable to find the thread. >> >> Following the wiki reconstruction workflow, I ran yesterday the recon-all -all process. The process finished without errors. So the next step was to check the talairach transform. >> Now, according to the FsTutorial/Talairach I have to make some adjustments and then run the whole process again. That is all nice and good. >> >> However, why is it, that the talairach volume is always, no matter which subject I look at, is always significantly smaller than the target volume? Is that normal? Has it to be fixed? Or is that just normal? Attached find one example. >> As I am using freesurfer for the first time, I am really unsure about that. >> >> Thanks for the help, >> Maria >> >> > > > 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. > > > -- Maria J. Felber, PhD Student Max Planck Institute for Human Cognitive and Brain Sciences Stephanstraße 1A 04103 Leipzig, Germany Phone: +49 341 9940-2465/2522 -- Maria J. Felber, PhD Student Max Planck Institute for Human Cognitive and Brain Sciences Stephanstraße 1A 04103 Leipzig, Germany Phone: +49 341 9940-2465/2522

1 0

Talairach transformation
by Maria Felber 13 Sep '11

13 Sep '11

Dear all, sorry, if that question came up before in the mailing list and I simple have been unable to find the thread. Following the wiki reconstruction workflow, I ran yesterday the recon-all -all process. The process finished without errors. So the next step was to check the talairach transform. Now, according to the FsTutorial/Talairach I have to make some adjustments and then run the whole process again. That is all nice and good. However, why is it, that the talairach volume is always, no matter which subject I look at, is always significantly smaller than the target volume? Is that normal? Has it to be fixed? Or is that just normal? Attached find one example. As I am using freesurfer for the first time, I am really unsure about that. Thanks for the help, Maria -- Maria J. Felber, PhD Student Max Planck Institute for Human Cognitive and Brain Sciences Stephanstraße 1A 04103 Leipzig, Germany Phone: +49 341 9940-2465/2522

2 3

Big Ventricles on v4.0
by Andrew Dumas 13 Sep '11

13 Sep '11

Hi everyone, I'm using a frozen version of FS v4.0 and I have a subject with rather large ventricles that don't get filled properly (surfaces and aseg are off) during recon. I can't really find a good fix in previous messages besides to upgrade FS, since it seems like the newer versions will handle it better. Are there any procedures I can follow with 4.0 to fix things (filling in the vents on wm.mgz, etc.)? Thanks! Andrew

2 1

Re: [Freesurfer] Recon-all Error: _FindFacePath
by Daniel Ferreira Padilla 13 Sep '11

13 Sep '11

Dear Bruce and experts, I would really appreciate if you can help me with my scans. I have successfully processed all my sample but got this same FindFacePath error with 2 subjects. I checked the brainmask and wm files and for me they look ok. So I do not know how can I fix this error. Thank you very much in advance ERROR: _FindFacePath: could not find path! Linux neuroimagen 2.6.35-30-generic #54-Ubuntu SMP Tue Jun 7 18:41:54 UTC 2011 x86_64 GNU/Linux recon-all -s GJ169_3D exited with ERRORS at Sat Aug 20 07:53:56 WEST 2011

3 2

Image normalization outcome and adjustment
by David Warren 12 Sep '11

12 Sep '11

Hi all, I'm attempting to use FreeSurfer 5.1 (version block below) to process some older T1 scans collected on a 1.5T magnet with a native voxel size of 0.9375x0.9375x1.5mm. Overall, FreeSurfer seems to deal with these scans just fine (resampling to 1 mm isotropic, etc.), but I'm a little concerned about the results of the built-in image normalization. The potential issue is illustrated in an image that I've posted to http://imgur.com/NiEVQ. The isolated plot to the right shows the intensity histogram that we've observed for more recent 3T T1 scans, with an obvious WM peak at 110 and a GM peak around 65. Meanwhile, each of the six plots to the left is the intensity histogram for one of our 1.5T scans. Although they are internally consistent (with the upper-leftmost example being something of an outlier), the GM peak is much closer to the WM peak, with a mode around 85. I'm wondering whether this reduced GM/WM separation is likely to influence the tissue segmentation process, and if so how I might best remedy the situation. At a guess, mri_normalize or mri_ca_normalize seem like useful functions, but I haven't found the documentation for either to be of much help. Any assistance or advice would be greatly appreciated! Thanks, Dave Warren -- Post-doctoral Fellow Neurology Department University of Iowa Hospitals and Clinics davideugenewarren(a)gmail.com -------- freesurfer-Linux-centos4_x86_64-stable-pub-v5.1.0 -------- Setting up environment for FreeSurfer/FS-FAST (and FSL) $Id: FreeSurferEnv.csh,v 1.79.2.1 2011/03/29 22:06:48 greve Exp $ FREESURFER_HOME /usr/freesurfer FSFAST_HOME /usr/freesurfer/fsfast FSF_OUTPUT_FORMAT nii.gz SUBJECTS_DIR /usr/freesurfer/subjects MNI_DIR /usr/freesurfer/mni FSL_DIR /usr/fsl

3 2

recon-all excluding parts of hippocampus
by Sheeva Azma 12 Sep '11

12 Sep '11

Dear Freesurfer List, I am using recon-all to measure cortical thickness in a population of Alzheimer's patients and I have noticed that in many patients the pial surface does not include the whole brain. This seems to be because there are hippocampal lesions that are dark spots in the image which throw off the segmentation. I have tried adding control points, but that is only for the white matter boundary, and so I was wondering if you could provide any insight as to how to correct the pial surface so that the entire hippocampus is included. I have attached an example image from tkmedit showing the pial and white boundaries. Thank you, Sheeva Azma

2 1

Project full annotation files from one subject to another
by mbrickho＠nmr.mgh.harvard.edu 12 Sep '11

12 Sep '11

Hi FS team, Is it possible to project an entire annotation file from one subject's surface to another? I haven't been able to find a command for this, but I might just not be looking hard enough. My intent is to project an annotation I've made on fsaverage to every member of a cohort of 500+ subjects, and then extract thickness data from each newly created annotation file. Currently, I do mri_annotation2label to split the original annotation into its individual labels. Then I project each label, one at a time, with mri_label2label. Then, I combine them back into an annotation with mri_label2annot for every subject. While this method works fine, it seems like it's more cumbersome than it has to be. Any help/advice would be appreciated. Thanks! Mike

2 1

mri_vol2vol
by Constantin Anastasopoulos 12 Sep '11

12 Sep '11

Hellom I coregistered manually a aparc+aseq.nii file to my DWI-space with tkregister2. I saved the .dat file and applied it later to the T1 image in order to apply the same spatial coregistration as to the aparc+aseg. For woth the aparc+aseg and the T1 I used the command mri_vol2vol --mov was the b0image --targ was the aparc+aseg and T1 respectively --o was the output (a .nii in both cases) --reg was the .dat file and at the and --inv. While the aparc+aseg registration to the b0 worked, the T1-registration gave a false result (altough it had the same dimensions and was in the same space with the aparc+aseg). What am I doing wrong? Is the mri_vol2vol command the wrong one? I want to apply the same changes as to the aparc+aseg, as they are saved in the .dat file. Thanks a lot, Constantin

2 3

some very basic Recon-all questions
by Martijn Cloos 10 Sep '11

10 Sep '11

Hi, When running something like: "recon-all -autorcon1 -i file1.nii.gz -s subject" it first converts the nifti file to the mgz format (orig.mgz, i think). In the proses the data appears to get converted to 8bit integers. Starting with float data how does this impact the analysis? It seems to me that the range selected in "orig.mgz" is not optimal and this appears to be resampled in T1.mgz. Are there ways to use a higher dynamic range, by forcing it to use floats or 16bit/32bit integers? Or should I just manually optimize/verify "orig.mgz" before running recon-all? Also, are there ways to use the original resolution instead of resampling to 1mm iso? Thanks! Martijn

2 1

bedpostx changes in past 2 years?
by swoodman＠nmr.mgh.harvard.edu 09 Sep '11

09 Sep '11

Hi all, We always run bedpostx in the developement (nmr-dev-env) environment, and after some digging around, we've noticed some differences in outputs between 2008 and now. Has anything changed with bedpostx in dev environment since 2008, that would effect data and/or analysis? There seem to be more log files in the bedpostx log directory in the sessions that were run more recently. So this is mainly why we are wondering if something has changed in the bedpostx program between October 2008 and October 2009 that affects the tractography data. Thank you! Sandra

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