Freesurfer

freesurfer@nmr.mgh.harvard.edu

4 participants
30076 discussions

mris_anatomical_stats question
by Morgan Hough 06 Aug '07

06 Aug '07

I am reporting thickness values from subjects native space constrained by mask labels using mris_anatomical_stats. I am currently getting the average of the unsmoothed estimates. I would like to be able to get smoothed values but the usage reports the smoothing option as -nsmooth <#> where the # is the # of smoothing iterations. If we are using 10mm FWHM smoothing for the group results can I estimate the number of smoothing iterations to obtain something equivalent? Would it be easier to use mri_surf2surf and create a smoothed thickness file for each subject? Thanks again. Cheers, -Morgan

1 0

Banks of the superior temporal sulcus question
by Morgan Hough 06 Aug '07

06 Aug '07

I was just wondering why the banks of the superior temporal sulcus was made into its own label on the Desikan template. Does this correspond (roughly) to a particular functional area. I didn't see anything in the paper about why it was chosen. Is it just easy to define given sulcal/gyral boundaries? Thanks in advance. Cheers, -Morgan

2 3

possible bug found in mri_surfcluster
by Robert Levy 06 Aug '07

06 Aug '07

Hello, I found a possible bug in mri_surfcluster, but it may be linked specifically to average7 and no other srcsubjects (but I haven't checked that). When I run mri_surfcluster with cluster simulation data, the size is reported at a seemingly arbitrary number, based on the fact that the #of vertices does not change. And the mgh output in null (a blank surface is displayed). However when I run mri_surfcluster with cluster simulation data, but specify srcsubject, it works correctly. This seems like a bug, because I think when surfcluster runs csd files it reads the srcsubject from the csd files, and no other srcsubject should be compatible with the data, therefore it must know the srcsubject in both usages, but in one case it works and the other it doesn't. Thanks, Rob * surflcluster with threshold of 1.3* # ClusterNo Max VtxMax Size(mm^2) TalX TalY TalZ NVtxs 1 -2.217 145639 *413.11* -6.4 38.4 5.0 *418* 2 -1.836 29811 141.08 -11.9 53.2 19.0 178 3 -1.532 16680 111.99 -11.1 19.1 22.6 178 4 -1.400 83520 17.42 -3.2 11.4 17.5 23 *surfcluster using csd files* # ClusterNo Max VtxMax Size(mm^2) TalX TalY TalZ CWP CWPLow CWPHi NVtxs 1 -2.217 145639 *275.40* -6.4 38.4 5.0 0.01190 0.01050 0.01330 *418* 2 -1.836 29811 94.06 -11.9 53.2 19.0 0.50970 0.50330 0.51610 178 3 -1.532 16680 74.66 -11.1 19.1 22.6 0.64580 0.63970 0.65190 178 4 -1.400 83520 11.61 -3.2 11.4 17.5 0.97920 0.97740 0.98100 23 *Apparently this is the right way to do it: with csd files, but specify srcsubj average7:* # ClusterNo Max VtxMax Size(mm^2) TalX TalY TalZ CWP CWPLow CWPHi NVtxs 1 -2.217 145639 *413.11* -6.4 38.4 5.0 0.00010 0.00000 0.00020 *418* 2 -1.836 29811 141.08 -11.9 53.2 19.0 0.27000 0.26430 0.27570 178 3 -1.532 16680 111.99 -11.1 19.1 22.6 0.40500 0.39870 0.41130 178 4 -1.400 83520 17.42 -3.2 11.4 17.5 0.96970 0.96750 0.97190 23 ------- freesurfer-Linux-centos4_x86_64-dev-20070802 -------- Setting up environment for FreeSurfer/FS-FAST (and FSL) FREESURFER_HOME /usr/local/freesurfer/dev FSFAST_HOME /usr/local/freesurfer/dev/fsfast FSF_OUTPUT_FORMAT nii SUBJECTS_DIR /space/ventzl/1/users/SUBJECTS_DIR MNI_DIR /usr/local/freesurfer/dev/mni FSL_DIR /usr/local/freesurfer/dev/fsl # PWD: /space/ventzl/1/users/SUBJECTS_DIR mri_surfcluster \ --src lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.mgh \ --csd ASevASc_8s_p05_z_lhMCSIM01-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM02-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM03-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM04-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM05-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM06-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM07-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM08-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM09-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM10-BwGroupError5.csd \ --sum lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.sum \ --ocp lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.mgh \ # overlay shows no clusters (tried at 0.5 as well in case it is a binary map) tksurfer average7 lh inflated -overlay lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.mgh mri_surfcluster \ --really-use-average7 \ --src lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.mgh \ --csd ASevASc_8s_p05_z_lhMCSIM01-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM02-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM03-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM04-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM05-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM06-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM07-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM08-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM09-BwGroupError5.csd \ --csd ASevASc_8s_p05_z_lhMCSIM10-BwGroupError5.csd \ --sum lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.sum \ --ocp lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.mgh \ --srcsubj average7 # overlay does show the clusters tksurfer average7 lh inflated -overlay lh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM_lh.mgh -- Robert P. Levy, B.A. Research Assistant, Manoach Lab Massachusetts General Hospital Charlestown Navy Yard 149 13th St., Room 2611 Charlestown, MA 02129 email: levy(a)nmr.mgh.harvard.edu phone: 617-726-1908 fax: 617-726-4078 http://nmr.mgh.harvard.edu/manoachlab

2 1

questions
by Paul Aparicio 06 Aug '07

06 Aug '07

Hello All, I have 2 questions I hope someone can help me with. 1) Is it possible to plot on a surface a set of arbitrary points in the coordinate frame of an anatomical volume? In other words if I have a set of x,y, and z points (not voxel indices: e.g. sub-voxel coordinates) can I plot them onto a surface? 2) Is there a way to view the slice lines in a volume and a surface? Can I see the anatomical location of the boundaries of the functional slices on the surface representation? Thanks paul

3 2

optseq2 question about fixing null intervals
by Denis R 06 Aug '07

06 Aug '07

Hello, We're designing a task that has two pairs of events (4 events total). Each pair of events has a pair of allowable null intervals associated with it. Is it possible to use optseq2 to create such a paradigm? Is it appropriate to use optseq2 in this case at all given these constraints?

2 1

How to do a surface-based group analysis with single-subject SPM5 contrast images
by Johannes Klackl 06 Aug '07

06 Aug '07

Dear all, I'm currently trying to conduct a surface-based group analysis using 1st level contrast images from SPM5. I wanted to ask if somebody could tell me if what I'm doing is OK. I would relly appreciate your help! I will now describe the procedure: The functionals of each subject were realigned and unwarped using SPM5 preprocessing commands and then entered into an SPM 1st level designs. Additionally, of course, cortical models from all subjects were created using Freesurfer. I used spmregister to realign the contrast images (con_0001.img) with the Freesurfer anatomicals. spmregister --s v01 --mov ../f2/auS1_v01_06LOC_x001.img --reg v01_register.dat Then I used mri_vol2surf (with the resulting registration file from spmregister) to bring the volumetric information from the contrast images onto each subject's individual surface. mri_vol2surf --hemi lh --src con_0001.img --src_type analyze --srcreg v01_register.dat --out lh.con_0001.mgh --out_type mgh --projfrac 0.5 --noreshape After that, I used mri_surf2surf with ico as the target to bring all the functionals from the individual subject surface onto a common surface. mri_surf2surf --hemi lh --srcsubject v01 --srcsurfval lh.con_0001.mgh --src_type mgh --trgsubject ico --trgicoorder 7 --trgsurfval lh.ico.con_0001.mgh --trg_type mgh I was thinking about smoothing here, but I didn't do this yet. Then, I used mri_concat to concatenate the functionals from all subjects into one file and mri_glmfit to do the statistics. Thanks, Johannes Klackl, University of Salzburg

2 1

Re: [Freesurfer] Fake thickness map 2
by Inês Souta 06 Aug '07

06 Aug '07

And after creating the mgh volume, how can I read it? Is it possible to open it in tksurfer? Should I load it as an overlay? I've used one of my already pre-processed subjects, created a 'vol' with the same number of vertices of one of the hemisferes, and saved the output of save_mgh file as an rh.thickness.mgh file. However, when I don't seem to be able to open it in tksurfer... Thanks Ines 2007/8/6, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu>: > > they don't matter. For a real volume they would be the ras2vox transform > and the mr parameters. For your purposes they can be any 4x4 and you can > leave the mri_parms one out (it will detect and fill it in with zeros). > > Send these to the list so other people can answer! > > Bruce > On > Mon, 6 Aug 2007, Inês Souta wrote: > > > Hi again, > > > > I still don't understand what all the input objects of save_mgh function > > stand for. 'vol' is the vector of thickness values, 'fname' is the name > of > > the output .mgh file, and what about M and <mr_parms>? What do they > stand > > for and how should I define them? > > > > Thanks again, > > > > Ines > > > > 2007/7/31, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu>: > >> > >> Hi Ines, > >> > >> use the mgh route. The "vol" is actually the vector of thickness > values, > >> with one entry per vertex on the surface. > >> > >> cheers, > >> Bruce > >> On Tue, 31 Jul 2007, Inês Souta > >> wrote: > >> > >>> Hello, > >>> > >>> I was trying to create a cortical thickness map with constant values > in > >>> order to perform some tests. At first I tried using mris_convert. I > >>> converted my thickness file into ascii: > >>> > >>> mris_convert lh.thickness lh.thickess.asc > >>> > >>> then I used matlab to change all the values in the last column to 2,0. > I > >>> tried to convert the ascii file back to surface file by doing > >>> > >>> mris_convert lh.thicness.asc lh.thickness > >>> > >>> but I got "Segmentation fault". How can I convert may ascii file into > a > >>> surface one? > >>> > >>> After failling to acomplish this, I tried using the save_mgh and > >> load_mgh > >>> file, nevertheless I'm not quite sure which input values are needed. > >> Where > >>> can I put my fake values in the save_mgh file? And what is "vol"? > >>> > >>> Thanks, > >>> > >>> Ines > >>> > >

2 3

Fake thickness map 2
by Inês Souta 06 Aug '07

06 Aug '07

---------- Forwarded message ---------- From: Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> Date: 06/08/2007 14:23 Subject: Re: [Freesurfer] Fake thickness map 2 To: Inês Souta <ines.souta(a)gmail.com> they don't matter. For a real volume they would be the ras2vox transform and the mr parameters. For your purposes they can be any 4x4 and you can leave the mri_parms one out (it will detect and fill it in with zeros). Send these to the list so other people can answer! Bruce On Mon, 6 Aug 2007, Inês Souta wrote: > Hi again, > > I still don't understand what all the input objects of save_mgh function > stand for. 'vol' is the vector of thickness values, 'fname' is the name of > the output .mgh file, and what about M and <mr_parms>? What do they stand > for and how should I define them? > > Thanks again, > > Ines > > 2007/7/31, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu>: >> >> Hi Ines, >> >> use the mgh route. The "vol" is actually the vector of thickness values, >> with one entry per vertex on the surface. >> >> cheers, >> Bruce >> On Tue, 31 Jul 2007, Inês Souta >> wrote: >> >>> Hello, >>> >>> I was trying to create a cortical thickness map with constant values in >>> order to perform some tests. At first I tried using mris_convert. I >>> converted my thickness file into ascii: >>> >>> mris_convert lh.thickness lh.thickess.asc >>> >>> then I used matlab to change all the values in the last column to 2,0. I >>> tried to convert the ascii file back to surface file by doing >>> >>> mris_convert lh.thicness.asc lh.thickness >>> >>> but I got "Segmentation fault". How can I convert may ascii file into a >>> surface one? >>> >>> After failling to acomplish this, I tried using the save_mgh and >> load_mgh >>> file, nevertheless I'm not quite sure which input values are needed. >> Where >>> can I put my fake values in the save_mgh file? And what is "vol"? >>> >>> Thanks, >>> >>> Ines >>> >

1 0

Re: [Freesurfer] show the results
by Bruce Fischl 03 Aug '07

03 Aug '07

look in /Users/lning/subjects/MGH_Ruphert/mri and /Users/lning/subjects/MGH_Ruphert/surf and see if there is any processed data there On Fri, 3 Aug 2007, Ning Liu wrote: > I think so. It is someone's data. He had created the surfers. I just wanted > to have a look at it. > > On 8/3/07, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> wrote: >> >> did you run recon-all on this subject? >> >> On Fri, 3 Aug 2007, Ning Liu wrote: >> >>> I copied some subjects' data to my computer, but i cannot run the surfer >>> >>> here is the command line: >>> >>> tksurfer MGH_Ruphert lh inflated >>> >>> screen output >>> >>> surfer: current subjects dir: /Users/lning/subjects >>> surfer: not in "scripts" dir ==> using cwd for session root >>> surfer: session root data dir ($session) set to: >>> surfer: /Users/lning/subjects/MGH_Ruphert >>> surfer: ### Could not find anatomical header information >>> >>> >> > > > >

1 0

starting midway in the pipeline
by Brianna Schuyler 03 Aug '07

03 Aug '07

Hello, I would really like to use a freesurfer surface to display non-human primate data that I have already put into a standard space and segmented, but I am having trouble finding information on how to skip many of the initial steps to this point. Also, I have seen references to a "-monkey" option for some steps, but I haven't been able to find any documentation on that either. Thanks for any help, Brianna Schuyler

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