Freesurfer

freesurfer@nmr.mgh.harvard.edu

18 participants
30131 discussions

tksurfer ftimepoint variable
by Yigal Agam 14 Nov '08

14 Nov '08

Hi, I'm trying to use a TCL script that runs through every time point in a surface. I'm using the following commands: set gaLinkedVars(ftimepoint) 0 (or 1, 2, etc) SendLinkedVarGroup overlay But tksurfer is always showing point 0 (doesn't advance in time). Similar TCL commands for fthresh and fslope work fine. Any ideas?

1 0

Antw: Re: [Freesurfer] absolute curvature values for vertex-wise comparison
by Carl Schultz 14 Nov '08

14 Nov '08

Hi Bruce, thank you very much. That' s what I looked for. Best regards, Christoph >>> Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> 13.11.08 14.43 Uhr >>> Hi Carl, you could read them into matlab, take the absolute value, and write them back out again if you want, then run a glm on that. load_mgh and save_mgh are the matlab i/o routines. cheers, Bruce On Thu, 13 Nov 2008, Carl Schultz wrote: > Dear all, > > for a vertex-wise comparison of the mean curvature is there a > possibility to change the signed curvature values from the preprocessed > curv.mgh to unsigned (absolute) values for the glm? > > Thanks in advance, > > Christoph > ____________________ > Universitätsklinikum Jena > Körperschaft des öffentlichen Rechts und Teilkörperschaft der > Friedrich-Schiller-Universität Jena > Bachstraße 18, 07743 Jena > Verwaltungsratsvorsitzender: Prof. Dr. Walter Bauer-Wabnegg; > Medizinischer Vorstand: Prof. Dr. Klaus Höffken; > Wissenschaftlicher Vorstand: Prof. Dr. Klaus Benndorf; Kaufmännischer > Vorstand und Sprecher des Klinikumsvorstandes Rudolf Kruse > Bankverbindung: Sparkasse Jena; BLZ: 830 530 30; Kto.: 221; > Gerichtsstand Jena > Steuernummer: 161/144/02978; USt.-IdNr. : DE 150545777 > > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > ____________________ Universitätsklinikum Jena Körperschaft des öffentlichen Rechts und Teilkörperschaft der Friedrich-Schiller-Universität Jena Bachstraße 18, 07743 Jena Verwaltungsratsvorsitzender: Prof. Dr. Walter Bauer-Wabnegg; Medizinischer Vorstand: Prof. Dr. Klaus Höffken; Wissenschaftlicher Vorstand: Prof. Dr. Klaus Benndorf; Kaufmännischer Vorstand und Sprecher des Klinikumsvorstandes Rudolf Kruse Bankverbindung: Sparkasse Jena; BLZ: 830 530 30; Kto.: 221; Gerichtsstand Jena Steuernummer: 161/144/02978; USt.-IdNr. : DE 150545777

1 0

Antw: Re: [Freesurfer] absolute curvature values for vertex-wise comparison
by Carl Schultz 14 Nov '08

14 Nov '08

Dear Rudolph, thank you very much for your reply. I used the load_mgh command in matlab for the preprocessed curv.mgh to create the absolute_curvature _values.mgh, which was successful. Best regards, Christoph >>> Rudolph Pienaar <rudolph(a)nmr.mgh.harvard.edu> 13.11.08 22.44 Uhr >>> On Thursday 13 November 2008 08:23:08 Carl Schultz wrote: > Dear all, > > for a vertex-wise comparison of the mean curvature is there a > possibility to change the signed curvature values from the preprocessed > curv.mgh to unsigned (absolute) values for the glm? > The newest (i.e. dev) version of 'mris_calc' will do this: $>mris_calc rh.curv abs This will run an 'abs' (absolute value) function on each element in 'rh.curv'. Output is saved by default to 'out.crv'. You can specify a more meaningful output with '--output <someFile>', so $>mris_calc --output rh.curv.abs rh.curv abs Since this is only in the 'dev' tree, let me know if you'd like a binary. -=R -- Rudolph Pienaar, M.Eng, D.Eng / email: rudolph(a)nmr.mgh.harvard.edu MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging 149 (2301) 13th Street, Charlestown, MA 02129 USA _______________________________________________ Freesurfer mailing list Freesurfer(a)nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ____________________ Universitätsklinikum Jena Körperschaft des öffentlichen Rechts und Teilkörperschaft der Friedrich-Schiller-Universität Jena Bachstraße 18, 07743 Jena Verwaltungsratsvorsitzender: Prof. Dr. Walter Bauer-Wabnegg; Medizinischer Vorstand: Prof. Dr. Klaus Höffken; Wissenschaftlicher Vorstand: Prof. Dr. Klaus Benndorf; Kaufmännischer Vorstand und Sprecher des Klinikumsvorstandes Rudolf Kruse Bankverbindung: Sparkasse Jena; BLZ: 830 530 30; Kto.: 221; Gerichtsstand Jena Steuernummer: 161/144/02978; USt.-IdNr. : DE 150545777

1 0

absolute curvature values for vertex-wise comparison
by Carl Schultz 13 Nov '08

13 Nov '08

Dear all, for a vertex-wise comparison of the mean curvature is there a possibility to change the signed curvature values from the preprocessed curv.mgh to unsigned (absolute) values for the glm? Thanks in advance, Christoph ____________________ Universitätsklinikum Jena Körperschaft des öffentlichen Rechts und Teilkörperschaft der Friedrich-Schiller-Universität Jena Bachstraße 18, 07743 Jena Verwaltungsratsvorsitzender: Prof. Dr. Walter Bauer-Wabnegg; Medizinischer Vorstand: Prof. Dr. Klaus Höffken; Wissenschaftlicher Vorstand: Prof. Dr. Klaus Benndorf; Kaufmännischer Vorstand und Sprecher des Klinikumsvorstandes Rudolf Kruse Bankverbindung: Sparkasse Jena; BLZ: 830 530 30; Kto.: 221; Gerichtsstand Jena Steuernummer: 161/144/02978; USt.-IdNr. : DE 150545777

3 2

A problem on groupanalysis
by Zhangyuanchao 13 Nov '08

13 Nov '08

Dear Mr or Ms, I have read the FreeSurfer Tutorial: Group Analysis, but there are still something I do not quite understand. I want to make comparison of lgi between two groups, namely, normal control group and patient group. I want to regress out gender and age.For example,if I have 4 subjects: nc1 male age=30,nc2 female age=31,patient1 male age=32, patient2 female age=30, how should I create the FSGD file?How should I design the constrast vector? What kind of test does Freesurfer use,F-test or t-test to detect difference region of thickness or lgi on the surface? What should I do if I want to use t-test to detect the difference between two groups? How should I set the option of the command mri_glmfit --sim nulltype nsim thresh csdbasename if I want to set the p-value threshold 0.05? Sorry for the trouble! Thank you very much! Yuanchao Zhang --------------------------------- 雅虎邮箱，您的终生邮箱！

2 1

Qdec problem?
by Martin Kavec 13 Nov '08

13 Nov '08

Hi, I am trying to run a group analysis on cortical thickness between patients and controls. In the qdec.dat I set a MMSE for all the controls 0.0. Now, whenever I try to covariate thickness against age and MMSE, the qdec complains that there is no spectrum in MMSE values. I tried to set MMSE for one of controls to 0.1 and the analysis goes fine. After thinking for a while, I started to wonder whether I shouldn't calculate thickness difference first, but I am not finding the receipt on the web anymore. Thanks in advance, Martin

2 1

mri_convert --conform spatial changes
by Arthur Mikhno 13 Nov '08

13 Nov '08

Hi, I am trying to convert the ribbon (amongst other mgz files) from a freesurfer run to the native space ANALYZE file I initially imported into freesurfer. I need to do this to create a cortical ribbon mask as a pre-processing step for normalization (with another package). However, every mgz file in the mri/ folder, after conversion, has a slightly different spatial distribution after conversion. Although, when I convert 001.mgz and rawavg.mgz the resultant image matches the image I imported exactly. By spatial distribution I mean, minor cortical fold shifts (up down, left right) and individual minor rotations in the cortical folds, some minor internal translations. I think I tracked the problem down to the conform step, when rawavg.mgz gets conformed (mri_convert --conform) to 1x1x1mm and saved as orig.mgz. I need to ensure a spatially correct cortical ribbon in the native space data. Is there any way to un-conform ? or some way to say, convert orig.mgz BACK to rawavg.mgz? For reference, this is the command I am using to convert .mgz to ANALYZE (.hdr, .img): mri_convert --out_orientation LSP -rl native_mri.img orig.mgz converted_native_mri.img I would really appreciate any help as this is my LAST roadblock! Thanks in advance, -Arthur

2 1

optseq2 question
by Hongchuan Zhang 13 Nov '08

13 Nov '08

Hi, list, I am new to optseq2 so I really need advices to my questions. We have an experiment of 2 factors with 36 trials for each factor, and another 24 null trials. So in total we have 96 trials. We also have 3 random jitters which make our trials could last for either 4.6, 5 or 5.4 seconds. The total session time is thus 480 seconds. Since our TR is 2 seconds, the ntp = 240. My questions are: (1) How can I set the psdwin? My intuition is that dPSD should be divided by both TR and trial duration. That means we can only have a dPSD = 0.2. Would that be too small? I observed that by doing this the largest psdmax is only 2 seconds. (2) What should I do if I want to break the session into two parts? The previous optseq version seems to allow one specify the runs for each sesssion. But I did not find this option in optseq2. Can I just use half the trials to get two sequences and use them for the two parts? Thanks! Your suggestion is appreciated. Hongchuan 2008-11-03 Hongchuan Zhang, Ph.D. Postdoctoral Research Associate Developmental Cognitive Neuroscience Laboratory Northwestern University 2240 North Campus Drive Frances Searle Building, Room 2-356 Evanston, Illinois, 60208 phone: (847)467-1549 e-mail: hongchuan-zhang at northwestern.edu

2 1

Re: [Freesurfer] tksurfer: Load GDF command line?]
by Pratap Kunwar 13 Nov '08

13 Nov '08

HI Tori, I don't think fsgd option is available on the tksurfer or tkmedit command line. It would have nice to load fsgd file on command line instead of loading from tksurfer gui. > Hi All, > I was wondering if there was a flag or command in tksurfer that would > allow me to load a group descriptor file (y.fsgd) using the command > line. I looked through the tutorial and couldn't seem to find anything. I > only found resources for loading a timecourse. This isn't the same thing > right? Any help would be appreciated! they are not same. > Thanks, > Tori > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > >

1 0

Functionality of autorecon2-wm
by Ryan Scotton 13 Nov '08

13 Nov '08

Hi all, My team has recently completed a few hundred FreeSurfer runs on young (2-4 year old) children. For the past month or so, we have been using various editing techniques in order to fix different problems with the white matter volumes (these were to be expected considering the ages of our subjects). Since our goal is to get accurate measurements of cortical thickness, this is a crucial step. A few of the images simply stop including the white matter at a certain point, always always toward the posterior/occipital region of the brain. We devised an approach to solving this problem that involves converting formats and orientations, loading the brainmask and white matter volume into a manual brain segmentation editing tool, and using a 'snake tool' that fills in the white matter that was not captured by FreeSurfer. We have ensured that the label we used to fill this in is the same as the label used for manual editing within tkmedit. HOWEVER, after rerunning FreeSurfer with this edited volume (using -autorecon2-wm), the surfaces do not seem to have changed at all. Am I missing something here? I thought that if white matter edits were made that filled in gaps in the original wm.mgz volume, FreeSurfer would adjust the the surfaces accordingly. I've attached two photos to give you an idea of what I'm experiencing. Thanks for any advice on this predicament. 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