Freesurfer June 2011

freesurfer@nmr.mgh.harvard.edu

118 participants
168 discussions

Thalamocortical Connectivity
by Kamal Abuhassan 08 Jun '11

08 Jun '11

Hi My PhD project focuses on developing a realistic thalamocortical model of Alzheimer's disease (AD). One of the major components of my project is to utilize DTI data from AD patients to build the macro-connectivity of the model. Currently, I'm working on FSL + FreeSurfer + CAMINO to process the DTI data. I'm sending this message to verify my approach in determining the regions of interest and to ask other critical questions. The approach A) Cortical Segmentation 1- First, I've segmented the T1-weighted image with recon-all command. 2- converted the cortical segmentation image (wmparc.mgz) into the Nifti format using mri_convert. B) Registration of Segmentation and Diffusion Weighted Image 1- The brain region was extracted from the b0 image using BET command 2- The output of the BET tool was used for the registration with the T1 image using FLIRT tool. 3- Obtain the inverse transformation matrix using convert_xfm in FSL 4- Apply the inverse matrix over the segmentation image with the 'Interpolation Method' as 'Nearest Neighbour' (using FLIRT). 5- Get the labels of Thalamus and cortical areas from usr/local/FreeSurfer/FreeSurferColorLUT.txt. (for cortical areas, I've considered cortical labels that started with 'wm-' but not 'ctx-'. Is that correct?) 6- Apply fslmath to pull out the appropriate ROI(s) 7- Apply Fiber Tracking tool to get fiber data. Questions: - Can you please advise if I'm doing right in outlining the ROI(s) ? - How can I differentiate between Thalamocortical and corticothalamic fibers ? - How can I get corticocortical fibers ? Is it right to get two cortical labels (source & destination) and find the pathways between them ? I will highly appreciate any advice/suggestions. Many Thanks Kamal.

1 0

subcortical mask to MNI152 space
by Elisa Golfinopoulos 07 Jun '11

07 Jun '11

Hi, I would like to normalize the subcort.mask.mgz to MNI152 space. I have attempted to use the command: mri_vol2vol --mov $FREESURFER_HOME/subjects/ fsaverage/mri.2mm/subcort.mask.mgz --targ $FSLDIR/data/standard/MNI152_T1_2mm.nii.gz --o ./subcort.mask152.nii --reg $FREESURFER_HOME/average/mni152.register.dat but when I check the registration with the MNI152 template I find that the 152 normalized subcortical mask does not align with the template. The problem is somewhat resolved when I specify a 180 degree rotation along the x axis (though there still appears to be some translation error). However, I feel uncomfortable using rotations/translations without understanding why the registration is incorrect in the first place. In fact, this problem is making me question whether a transformation is really necessary since, technically, both the mask and the target are in MNI space and the registration of the original MNI 305 subcortical mask with the MNI152 template looks more accurate than the 152 normalized mask. Any help that you can offer would be greatly appreciated. Cheers, Elisa -- Elisa Golfinopoulos Speech Laboratory Cognitive and Neural Systems Boston University Phone: 617.353.6181

2 1

wmparc.stats do not contain ctx colume after recon-all
by Ayaz, Muhammad 07 Jun '11

07 Jun '11

Dear All, After running recon-all I noticed wmparc.stats only contains wm volume and there is no ctx volume and subcortical volumes. Recon-all finished without any error. Thanks Ayaz The information contained in this message and any attachments is intended only for the use of the individual or entity to which it is addressed, and may contain information that is PRIVILEGED, CONFIDENTIAL, and exempt from disclosure under applicable law. If you are not the intended recipient, you are prohibited from copying, distributing, or using the information. Please contact the sender immediately by return e-mail and delete the original message from your system.

2 1

number of MPRAGES to use (3T - 32 channel 1mm ISO data)
by Ariel Brown 07 Jun '11

07 Jun '11

Hi All, We've collected 2 MPRAGEs per subject for a study. This is on a Siemens Tim Trio 3T with a 32-channel head coil, 1mm isotropic data. The SNR looks pretty good as expected. I've read in the freesurfer archive that it is not necessary to use two MPRAGEs with this high quality data, but I'm wondering if there is a drawback to using two since we've already got the data on everyone. I've compared the surface reconstruction and segmentation for a few subjects comparing one vs. two MPRAGEs and there is very little difference that I can see with my naked eye, except that for one subject there were more surface corrections needed for the subject without the two averaged together as opposed to just using the first image series. So, I guess what I'm asking is given this quality of data, would one recommend systematically using both MPRAGEs since I already have the data, using the first MPRAGE alone to decrease chances of blurring, or taking an extra step and trying to figure out for each individual which has less motion and using that one? Honestly, the last option does not seem to be totally realistic because there is very little motion in any of these images that appears subjectively. Thanks for your help. Ariel --------------- Ariel Brown, PhD Research Fellow Department of Psychiatry Mass General Hospital & Harvard Medical School Visiting Scientist Department of Brain & Cognitive Sciences Massachusetts Institute of Technology

2 1

Re: [Freesurfer] Middle line issue
by Wayne Su 07 Jun '11

07 Jun '11

> Hello, > I have a subject with very big ventricle. I managed to edit wm.mask and > aseg.mgz, and rerun the recon-all autorecon2, but the middle line separating > left/right ventricles is still not right (attached). Is there any way to > correct it? I¹m using FS 5.1. Thanks. > > Wayne >> >

1 0

wmparc.mgz
by Ayaz, Muhammad 06 Jun '11

06 Jun '11

Dear All, After running recon-all I noticed wmparc.mgz only contains wm volume and there is no ctx volume and subcortical volumes. Recon-all finished without any error. Thanks Ayaz The information contained in this message and any attachments is intended only for the use of the individual or entity to which it is addressed, and may contain information that is PRIVILEGED, CONFIDENTIAL, and exempt from disclosure under applicable law. If you are not the intended recipient, you are prohibited from copying, distributing, or using the information. Please contact the sender immediately by return e-mail and delete the original message from your system.

1 0

Creating DLPFC label
by Allison R. Kaup 06 Jun '11

06 Jun '11

Hello! I am attempting to create a DLPFC ROI, based on methods described in : Yendiki et al. (2010). Multi-site characterization of an fMRI working memory paradigm: Reliability ofactivation indices. Neuroimage. 53 (1), 119-131. Those authors state they merged existing parcellation labels, and then made a "coronal cut at Talairach coordinate y=26." I have successfully merged the existing labels, but am having trouble figuring out a good way to do the cut. I have tried using the "Make closed path" and "Custom fill" options in tksurfer, but am having trouble getting those to make a clean cut right at y=26. I'd appreciate any suggestions! Thank you, Allison

3 3

tracula tutorial data
by GREGORY R KIRK 06 Jun '11

06 Jun '11

can i get an update on the eta for the tracula associated tutorial data sets greg

1 0

MEF segmentation: FLASH_atlas.gca
by Auer, Tibor MD. Ph.D. 06 Jun '11

06 Jun '11

Dear Freesurferers, I am trying to process multi-echo FLASH (MEF) images with FS. I have read in FSWiki, that the FLASH_atlas.gca was built precisely for MEF data acquired with FA=5 and FA=30. However, when I have looked after FLASH_atlas.gca in the FS installation, I have found none. Is it possible to obtain it somehow? Thanks, Auer, Tibor M.D. Ph.D. Biomedizinische NMR Forschungs GmbH am Max-Planck Institut für Biophysikalische Chemie Am Fassberg 11 37077 Göttingen Germany Phone/Work: +49-(0)551-201-1725 Mobile: +49-(0)176-8012-7921 E-Mail: tauer(a)gwdg.de <mailto:tibor.auer@gmail.com>

2 3

cerebellum atlas template
by Anthony Dick 06 Jun '11

06 Jun '11

I didn't get a response the last post, so I'm going to try one more time! Hello all, We are looking to develop a cerebellum atlas training set based on the Schmahmann et al (2000) parcellation so we can automatically segment the cerebellum based on this atlas. Does this exist already? I know Makris and colleagues had done something similar to this, but I am not sure if it has been developed for Freesurfer. Anthony -- Anthony Steven Dick, Ph.D. Assistant Professor Department of Psychology Florida International University Modesto A. Maidique Campus DM 296B 11200 S.W. 8th Street Miami, FL 33199 Phone: 305-348-4202 (try this first) Lab Phone: 305-348-9057 or 305-348-9055 Fax: 305-348-3879 Email: adick(a)fiu.edu Webpage: http://www.fiu.edu/~adick

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 June 2011