Freesurfer

freesurfer@nmr.mgh.harvard.edu

19 participants
30088 discussions

acquisition in 1 session vs 2 sessions
by Daniel Ferreira 13 Nov '13

13 Nov '13

Thanks a lot Doug, We have followed your suggestion and now fMRI is the third sequence. Both patients and controls will have two sessions too. Best Daniel ---------------------------------------------------------------------- Message: 1 Date: Sat, 09 Nov 2013 12:50:39 -0500 From: Douglas Greve <greve(a)nmr.mgh.harvard.edu> Subject: Re: [Freesurfer] acquisition in 1 session vs 2 sessions To: freesurfer(a)nmr.mgh.harvard.edu Message-ID: <527E75EF.2070807(a)nmr.mgh.harvard.edu> Content-Type: text/plain; charset="iso-8859-1" I think the only thing you would have to worry about is the fMRI since the way the subject reacts to the task (or state of subject during rest) might change depending upon how long he/she has been in the scanner. Why not do the fMRI at #3? Then each population would be in the scanner for the same amount of time. If the T1 and FLAIR are bandwidth matched, I'd do them in the same session. The only other confound could be that your controls might move more in the last couple of scans since they will have been in the scanner for a while at that point. Depending on your patient population, there might be difference in movement anyway. There should not be any problems with analyzing the data regardless of how you slice it up. doug On 11/9/13 3:56 AM, Daniel Ferreira wrote: Dear experts, We are designing a study including a long MR protocol: 1) Localizer 2) Cal 3) T1-FSPGR 4) FLAIR 3D 5) DTI 1 6) DTI 2 7) fMRI Due to patients characteristics we were thinking on splitting the acquisition in two different sessions (session 1: sequences 1, 2, 3, 4; session 2: sequences 1, 2, 5, 6, 7). Do you think we should do the same for the healthy controls or we can scan them in just one single session? I wonder how system session-related artifacts could differentially affect the two kinds of protocol (spitted vs. single). I also wonder how FreeSurfer processing can be affected by this issues. Thanks a lot. Daniel Ferreira _______________________________________________ Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

1 0

(no subject)
by Gonzalo Rojas Costa 13 Nov '13

13 Nov '13

Hi: Which is the optimal parameters for a morphometric sequence (fspgr?) n a 1.5 T MR OPTIMA 450w GE MRI equipment ?... Because, we are using an fspgr 1mm isotropic sequence, but we got wrong defined borders in some structures... for example, hypocampus, cortex-wm... I think it could be because the GE images appears smoothed in comparison with the Siemens Avanto mprage images that we also use... Any comments ?.... Sincerely, Gonzalo Rojas Costa -------------------------------- Gonzalo Rojas Costa Laboratory for Advanced Medical Image Processing Department of Radiology Clínica las Condes Lo Fontecilla 441, Las Condes, Santiago, Chile. Tel: 56-2-2105170 Cel: 56-9-97771785 www.clc.cl

1 0

Re: [Freesurfer] bvecs and bvals don't have the same number of entries
by Anastasia Yendiki 12 Nov '13

12 Nov '13

The problem is that your original TLR108.bvec file is still in 3 rows, not 3 columns. You need to transpose the rows into columns. Yes, there's an example config file that comes with the 5.3 distribution and it's as usual in $FREESURFER_HOME/bin/dmrirc.example. It's also described on the wiki, where all the info reflects the latest version of the software. http://surfer.nmr.mgh.harvard.edu/fswiki/Tracula Please include freesurfer(a)nmr.mgh.harvard.edu in these emails so that the full thread is archived. Thanks! a.y On Tue, 12 Nov 2013, Laura Christine Anderson wrote: > Attached are bvecs and bvals files from dmri directory. Thanks for the > advice on Freesurfer 5.3- is there a sample config file for Tracula for > that version? > > Laura C. Anderson, B.A. > Graduate Student > Clinical & Developmental Psychology > University of Maryland > College Park, MD 20742 > BPS 0112 > > > > > On 11/12/13 11:52 AM, "Anastasia Yendiki" <ayendiki(a)nmr.mgh.harvard.edu> > wrote: > >> >> Thanks. What do these files look like? >> /export/data/brain1/RSTOM/TLR108/dmri/bvecs >> /export/data/brain1/RSTOM/TLR108/dmri/bvals >> >> These are the ones that the error message in trac-all.log is about, so >> something probably goes wrong with the conversion from your original >> files. >> >> Also, I recommend very strongly that you use the latest version of >> tracula >> (from freesurfer 5.3). It's improved on both speed and performance. >> >> >> On Tue, 12 Nov 2013, Laura Christine Anderson wrote: >> >>> Hi Anastasia, I have version 5.1. Attached are bvecs and bvals files >>> with >>> 3 and 1 columns respectively as well as the trac-all.log file. >>> >>> Thanks! >>> Laura >>> >>> Laura C. Anderson, B.A. >>> Graduate Student >>> Clinical & Developmental Psychology >>> University of Maryland >>> College Park, MD 20742 >>> BPS 0112 >>> >>> >>> >>> >>> On 11/11/13 9:00 PM, "Anastasia Yendiki" <ayendiki(a)nmr.mgh.harvard.edu> >>> wrote: >>> >>>> >>>> Hi Laura - First of all, which version of freesurfer do you have? It >>>> looks >>>> to me like you've based your config file on the old example file that >>>> came >>>> with version 5.1. >>>> >>>> Second, could you please convert your bvecs and bvals files to 3 >>>> columns >>>> and 1 column, respectively, then run trac-all and send the trac-all.log >>>> file that shows the errors that you get? >>>> >>>> Thanks! >>>> a.y >>>> >>>> On Mon, 11 Nov 2013, Laura Christine Anderson wrote: >>>> >>>>> Hi Anastasia, I am still struggling with this same error message, >>>>> despite >>>>> having rearranged the bval and bvec files in many different >>>>> configurations >>>>> (and making sure that each value in these files had the same number of >>>>> digits, which I am not sure even matters). I am attaching my most >>>>> recent >>>>> bval and bvec files as well as the config file to see if you have any >>>>> advice. >>>>> >>>>> Thanks so much for your help! >>>>> >>>>> Laura C. Anderson, B.A. >>>>> Graduate Student >>>>> Clinical & Developmental Psychology >>>>> University of Maryland >>>>> College Park, MD 20742 >>>>> BPS 0112 >>>>> >>>>> >>>>> >>>>> >>>>> On 10/15/13 11:17 AM, "Anastasia Yendiki" >>>>> <ayendiki(a)nmr.mgh.harvard.edu> >>>>> wrote: >>>>> >>>>>> >>>>>> Hi Laura - This means that there is only 1 line in each of the files, >>>>>> so >>>>>> the end-of-line characters aren't recognized. I'd make sure that the >>>>>> file >>>>>> is saved as plain text. >>>>>> >>>>>> a.y >>>>>> >>>>>> On Tue, 15 Oct 2013, Laura Christine Anderson wrote: >>>>>> >>>>>>> When I run "wc -l" on the bvec and bval files, I get "1" for both. >>>>>>> Running >>>>>>> just "wc" gives "1 279 3312" for bvec and "1 93 456" for bval >>>>>>> >>>>>>> Laura C. Anderson, B.A. >>>>>>> Graduate Student >>>>>>> Clinical & Developmental Psychology >>>>>>> University of Maryland >>>>>>> College Park, MD 20742 >>>>>>> BPS 0112 >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> On 10/14/13 5:18 PM, "Anastasia Yendiki" >>>>>>> <ayendiki(a)nmr.mgh.harvard.edu> >>>>>>> wrote: >>>>>>> >>>>>>>> >>>>>>>> Hi Laura - I don't think the error "bvecs and bvals don't have the >>>>>>>> same >>>>>>>> number of entries" should have to do with interspersed zero >>>>>>>> b-values. >>>>>>>> When >>>>>>>> you run "wc -l" on the bvecs and bvals files, what do you get? >>>>>>>> >>>>>>>> a.y >>>>>>>> >>>>>>>> On Mon, 14 Oct 2013, Laura Christine Anderson wrote: >>>>>>>> >>>>>>>>> Hi Anastasia, the bval/bvec files were created on a Mac and not >>>>>>>>> with >>>>>>>>> excel. I am not seeing the ^M character when I view the files in >>>>>>>>> unix >>>>>>>>> with >>>>>>>>> the more command or in gedit perhaps when I sent you the files, >>>>>>>>> something >>>>>>>>> became corrupted? I am wondering if my issue has something to do >>>>>>>>> with >>>>>>>>> the >>>>>>>>> fact that there are 3 sets of 30 rows with a b0 image before each >>>>>>>>> of >>>>>>>>> the 3 >>>>>>>>> sets. In the examples I have seen, the b0 images all seem to be at >>>>>>>>> the >>>>>>>>> beginning of the bvec and bval files. I am attaching a screenshot >>>>>>>>> of >>>>>>>>> the >>>>>>>>> bvec and bval files as I see them in case that is helpful. If you >>>>>>>>> would >>>>>>>>> rather me send the actual files again, what is the best way to do >>>>>>>>> that >>>>>>>>> without the formatting getting messed up? Before, I was using a >>>>>>>>> free >>>>>>>>> fileshare program. >>>>>>>>> >>>>>>>>> Thank you! >>>>>>>>> Laura >>>>>>>>> >>>>>>>>> Laura C. Anderson, B.A. >>>>>>>>> Graduate Student >>>>>>>>> Clinical & Developmental Psychology >>>>>>>>> University of Maryland >>>>>>>>> College Park, MD 20742 >>>>>>>>> BPS 0112 >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On 10/11/13 2:38 PM, "Anastasia Yendiki" >>>>>>>>> <ayendiki(a)nmr.mgh.harvard.edu> >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>> >>>>>>>>>> Hi Laura - Were these bval/bvec files created on windows or with >>>>>>>>>> excel >>>>>>>>>> by >>>>>>>>>> any chance? When you view them in unix (with the more command for >>>>>>>>>> example) >>>>>>>>>> you'll see an "^M" character instead of every end-of-line. Can >>>>>>>>>> you >>>>>>>>>> try >>>>>>>>>> editing these files in a unix editor to fix that? Let me know if >>>>>>>>>> this >>>>>>>>>> solves the problem. >>>>>>>>>> >>>>>>>>>> a.y >>>>>>>>>> >>>>>>>>>> On Fri, 11 Oct 2013, Laura Christine Anderson wrote: >>>>>>>>>> >>>>>>>>>>> Hi Anastasia, >>>>>>>>>>> Thanks so much for your help! Attached are bvec, bval and >>>>>>>>>>> trac-all.log >>>>>>>>>>> files. >>>>>>>>>>> >>>>>>>>>>> I have been stuck at one of the preprocessing steps (trac all) >>>>>>>>>>> for >>>>>>>>>>> some >>>>>>>>>>> time and keep getting the >>>>>>>>>>> error: bvecs and bvals don't have the same number of entries. I >>>>>>>>>>> saw >>>>>>>>>>> a >>>>>>>>>>> couple blogs online about >>>>>>>>>>> this, but did not see any resolution. I am wondering if it is >>>>>>>>>>> because >>>>>>>>>>> the B0 images are >>>>>>>>>>> interspersed? (see attached bvec file). If you have any >>>>>>>>>>> insight, I >>>>>>>>>>> would love some feedback >>>>>>>>>>> whenever you get the chance. >>>>>>>>>>> >>>>>>>>>>> Laura >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> 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. >>>>>>>>> >>>>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>> >>>>> >>> >>> > >

2 2

Lower age limit
by Victor Kovac 12 Nov '13

12 Nov '13

Hello Freesurfers, I am just wondering if there is a generally accepted lower age limit for running recon-all and getting reliable results? Lack of myelination in most children below the age of 3 has caused segmentation issues in our dataset. Is this common? And, if so, has anything been published attesting to this? Thanks! Victor

2 1

freesurfer upgrade
by Octavian Lie 12 Nov '13

12 Nov '13

Dear All, We use freesurfer in an automatic routine mainly for brain extraction/ parcellation schemes. I want to upgrade from 5.2 to 5.3 on a linux (ubuntu, 64bit) computer, but without surprises. So: 1. When extracting files/installing freesurfer tar.gz, would the update override the 5.2 files if I want to keep the same path, or I have to delete 5.2 first? I obviously want to preserve '/subjects' data already processed. 2. Freesurfer distribution contains some CUDA files. On the computer, I have CUDA 5.5 version. Do you see any incompatibility issues (5.2 worked ok in this respect). 3. Could you send me the T1 and T2 space MR protocols you use for assistance with getting pial recon optimized. Thank you, Octavian

2 1

Questions regarding stats file
by pfotiad＠nmr.mgh.harvard.edu 12 Nov '13

12 Nov '13

Hi, I am using FS v5.3 and I had two questions regarding the calculation of cortical thickness in the stats files: 1) Is the average cortical thickness of each structure (i.e. precuneus) calculated by averaging the cortical thickness values of each location in that structure? 2) Is the mean thickness calculated in the *h.aparc.stats (or in the other aparc stats files) calculated by averaging the cortical thickness values of all the structures in that hemisphere, or by using a weighted average of the cortical thicknesses of the structures included in that hemisphere? Thanks, Panos

2 4

recon all question
by Rashmi Singh 12 Nov '13

12 Nov '13

Hello FreeSurfer experts, I have a question regarding group analysis and running of recon all group wise. For example I have group A, B, and C with several subjects in each group, if want want to compare their cortical and subcortical measurements group wise, how do I run the recon all. Also can I compare the groups statistical within free surfer? Thank you! Rashmi

2 1

error using selxavg3-sess for resting state connectivity
by Scott Cameron Kolbe 12 Nov '13

12 Nov '13

Hi there We are performing a connectivity analysis using a seed mask we have drawn from a label on the average surface in tksurfer. We put the label into subject surface space using mri_label2label then made the label into a volume using mri_label2vol. We then extracted the seed timecourse using fcseed-config and fcseed-sess. We then have a text file containing the seed region timeseries in the fMRI data folder. We then ran mkanalysis-sess with the -notask flag and after the -taskreg flag we input the seed region timeseries text file. mkanalysis-sess -a connectivity -surface fsaverage lh -notask -taskreg Sess01/rest/001/connect_FEF2FEF.txt 1 -nuisreg Sess01/rest/001/ica_reg.txt -1 -fsd rest -TR 2.5 When we run selxavg3-sess after this we get an error after it starts matlab: ERROR: cannot find volume matching /data/project2/pMelb0036/om_2013/freesurfer/project/Sess01/rest/001/Sess01/rest/001/connect_FEF2FEF.txt ERROR: loading nonpar reg /data/project2/pMelb0036/om_2013/freesurfer/project/Sess01/rest/001/Sess01/rest/001/connect_FEF2FEF.txt ------------------------------------------ ERROR: fast_selxavg3() failed\n I'm not sure why it can't find the file as it's definitely there in the directory containing fmcpr.nii can anyone help? many thanks Scott

2 1

Automated batch ROI
by Wesley Burge 12 Nov '13

12 Nov '13

Dear Freesurfer experts, I am about to start an analysis where we will be drawing surface ROIs using the select_talairach_point option, making a label of the vertex selected, and dilating it a set number of times. Each subject has 19 ROIs and we have around 50 subjects. I was wondering if anyone knew of a way to use a script to automate this process. Thanks! -- Wesley Burge Behavioral Neuroscience Graduate Research Assistant UAB Dept of Psychology 1720 2nd Avenue South Birmingham, AL 35294 [image: Description: UAB Blazer]

2 1

localGI & Matlab runtime?
by Ryan Muetzel 12 Nov '13

12 Nov '13

Hello, I was wondering if anyone has compiled the localGI code with the matlab runtime & integrated with recon-all? We have a lot of data that we'd like to run through the localGI stream, however we have a limited number of matlab licenses. Are there other easy ways to "parallelize" this (i.e., run different subjects on different cores) without using up many matlab licenses? Thank you! All the best, Ryan

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