Freesurfer March 2009

freesurfer@nmr.mgh.harvard.edu

73 participants
100 discussions

FDR correction
by Lars M. Rimol 26 Mar '09

26 Mar '09

Hi, I have done an analysis involving three groups, so there are three pairwise comparisons across two hemispheres = 6 p-maps. I want to adjust for multiple comparisons (across the vertices), so I use FDR. But since FDR determines the threshold basd on the actual p-values, I get 6 different tresholds: comparison 1: lh and rh, 0.016 and 0.028 (I can choose .01) comparison 2: lh and rh, 0.01 and 0.001 (I can choose.001) comparison 3 lh and rh, 0.001 and 0.0001 (I can choose .0001) There are lots of significant vertices in comparison 1 and nothing significant, after correction, in comparison 3. Is there anything wrong with using different tresholds here, and concluding that in comparison 1 there were extensive differences between the groups, whereas in comparison 3 there were none? I'm not sure if this is a problem, but I'm afraid some reviewers might have an issue with it. Across the hemispheres, I can choose a conservative threshold which covers both hemispheres, i.e. lower than both the FDR-adjusted treshold for lh and rh. But between the comparisons the tresholds differ even more, by a factor of 10 and 100. And if I choose the most conservative of all the adjusted thresholds, I'm afraid that I'll make a type II error in comparison 1. >From what I understand, the adjusted threshold for comparison 3 is more conservative because of the actual empirical data (the distribution of p-values), so that's an empirical argument for using a more conservative threshold there. And: What if I pooled all thre p-maps (sig.mgh) and did an FDR on the whole thing, would that be a better approach? And does Freesurfer use the Benjamini algorithm, and if you do, can I use Tom Nichols' matlab function for FDR (http://www.sph.umich.edu/~nichols/FDR/FDR.m) for pooling all three p-maps? Thank you! -- yours, LMR

3 4

How to project fMRI activity of entire depth of cortex on to surface
by Wang, Xin 25 Mar '09

25 Mar '09

Hello, Freesurfer group, The mri_vol2surf only register the fMRI activity at a certain depth of cortex, as specisfied by -projecfrac, to the cortical surface. mri_surf2surf does the same thing. I want to identify any cortical region have activation during the task, therefore, I need to project fMRI activity of entire depth of cortex onto surface. Some regions may have activity in different depth of cortex, Ideally, the significent level of these regions reflect the peak of the activity cross entire depth or the sum of significense of entire depth. the --interp trilinear may do some toward my goal, but it elimit many activities compare to nearest. Thanks, Xin Wang

2 4

Monte Carlos for volume data?
by Dave Brohawn 25 Mar '09

25 Mar '09

Hello, Here is my bugr information, if you need. FREESURFER_HOME: /usr/local/freesurfer/stable4 Build stamp: freesurfer-Linux-centos4-stable-v4.2.0e-20090320 RedHat release: CentOS release 5.2 (Final) Kernel info: Linux 2.6.18-92.1.22.el5 i686 NMR Center info (/space/freesurfer exists): machine: seymour SUBJECTS_DIR: /autofs/space/ventzl_016/users/ PWD: /autofs/space/ventzl_016/users ssh seymour setenv SUBJECTS_DIR /autofs/space/ventzl_016/users/ cd /autofs/space/ventzl_016/users Can you tell me if monte carlo simulations can be used for volume data? Josh Roffman and I are curious if clusters seen in the volume will pass monte carlo simulations. I have been working closely with Kara Dyckman, and we couldn't find a precedent for monte carlos involving volume data completed in the Manoach group, a freesurfer example, or a mention of it in the archives. I used an mgh file from the mris_preproc command below to complete monte carlos for surface data on the left hemisphere. mris_preproc --fsgd /autofs/space/ventzl_016/users/fsgd/qdec.fsgd --target fsaverage --hemi lh --sf /autofs/space/ventzl_016/users/Subject_files/All-18 --a newEMerror-tpx --c ASevASc_6s --out /autofs/space/ventzl_016/users/All-18/preproc/lh.ASevASc.All-18.gnr.mgh I examined a log created from this preproc command, and saw that mri_vol2surf was completed for each subject in the group, followed by a concatenate command to create the mgh file. Does this mean that the volume data is taken into account in this mgh file? Here is one of the ten monte carlos I ran for the surface data. The other nine differ only in the MCSIM #. mri_glmfit --y $SUBJECTS_DIR/All-18/preproc/lh.ASevASc.All-18.gnr.mgh --fsgd $SUBJECTS_DIR/fsgd/All-18.fsgd --surf fsaverage lh --C $SUBJECTS_DIR/fsgd/All-18.mat --sim mc-full 1000 2 $SUBJECTS_DIR/All-18/monte_carlo_01/lh_ASevASc_6_rfx_p05_f_MCSIM1 --fwhm 14.41 If the volume data was processed in the preproc to make the mgh file for both hemispheres, then do these monte carlos also take volume data into account? Do I run a mri_volcluster (as opposed to surfcluster) to assess volume clusters and see if they pass monte carlos with what I already have, then? Thanks for taking the time to read this through. I know this email is on the long side. Dave Brohawn

1 0

Measuring cortical thickness with high resolution data
by Falk Lüsebrink 25 Mar '09

25 Mar '09

Hi Freesurfers, Im trying to evaluate the usefulness of high resolution scans acquired at 7T with an isometric voxel size of .6mm for the measurement of cortical thickness. The inhomogeneities are taken care of by dividing the scans with another scan of another sequence, so they are not an issue anymore. My problem is that Freesurfer usually conforms the voxel size to 1mm which is not desirable. I tried using the cm and noaseg flags for the recon-all process to avoid the conformation and to skip the subcortical segmentation, but another problem arises while using these flags. The error I receive is occurring after the WM Segmentation and states as follows: #-------------------------------------------- #...@# WM Segmentation Wed Mar 18 10:14:42 CET 2009 cp wm.mgz wm.seg.mgz mri_segment -keep brain.mgz wm.seg.mgz preserving editing changes in output volume... doing initial intensity segmentation... using local statistics to label ambiguous voxels... computing class statistics for intensity windows... WM (106.0): 106.9 +- 5.8 [80.0 --> 125.0] GM (69.0) : 66.3 +- 11.6 [30.0 --> 96.0] setting bottom of white matter range to 77.9 setting top of gray matter range to 89.4 doing initial intensity segmentation... using local statistics to label ambiguous voxels... using local geometry to label remaining ambiguous voxels... reclassifying voxels using Gaussian border classifier... removing voxels with positive offset direction... smoothing T1 volume with sigma = 0.250 removing 1-dimensional structures... thickening thin strands.... 20 segments, 4813 filled 10270 bright non-wm voxels segmented. 5589 diagonally connected voxels added... white matter segmentation took 2.4 minutes writing output to wm.seg.mgz... ERROR: mri_segment-MRIcheckVolDims: volume1 height=256 != volume2 height=320. The dimensions of the data Im trying to process is 320 x 320 x 224 and is changed to 320 x 320 x 320 while using the cm flag. It seems the mri_segment process cant handle any dimensions above 256 or Im missing another flag. Does someone has any ideas about that issue? Regards, Falk

1 0

No variation in white matter volume between groups as well as hemispheres
by Jürgen Hänggi 25 Mar '09

25 Mar '09

Dear FS experts We analyzed two groups of six subjects each and recognized that the mean total white matter volume (as reported in aseg) is identical between the two groups as well as between the two hemispheres: group1: mean/sd: lh 244.2/32.3 cm3; rh 244.2/32.3 cm3 group2: mean/sd: lh 244.2/32.3 cm3; rh 244.2/32.3 cm3 However, when looking at the individual subjects there is variation between the individuals as well as the left and the right hemisphere. Any ideas? Thanks in advance Regards Jürgen ---------------------------------------------------------------------------- Juergen Haenggi Ph.D. (Dr. des.) Division Neuropsychology Institute of Psychology University of Zurich Binzmuehlestrasse 14, PO Box 25 8050 Zurich, Switzerland 0041 44 635 73 97 (phone office) 0041 76 445 86 84 (phone mobile) 0041 44 635 74 09 (fax office) BIN 4.D.04 (office room number) j.haenggi[at]psychologie.uzh.ch (email) http://www.psychologie.uzh.ch/neuropsy/ (website) http://www.juergenhaenggi.ch (private website) This e-mail (and any attachment/s) contains confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorised copying, disclosure or distribution of the material in this e-mail is strictly forbidden. ----------------------------------------------------------------------------

1 0

New Tutorial Attachments
by Jeff Sadino 24 Mar '09

24 Mar '09

Hello, I wanted to review the FreeSurfer attachments on the wiki's tutorial page http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial Whenever I try to click on an attachment, it brings me to a page and says: Attached FilesNo attachments stored for FsTutorial/April2009CourseSchedule You are not allowed to attach a file to this page. I see the page was recently updated for the new class. Are the powerpoints still available to view somewhere? Mahalo, Jeff Sadino _________________________________________________________________ Hotmail® is up to 70% faster. Now good news travels really fast. http://windowslive.com/online/hotmail?ocid=TXT_TAGLM_WL_HM_70faster_032009

2 1

corpus callosum mask
by Al-Radaideh Ali 24 Mar '09

24 Mar '09

Dear FreeSurfer Experts, I wonder if Freesurfer saves the mask of corpus callosum or not. If so, could you please tell me where it is saved or how can I get a cc mask? Many thanks Ali This message has been checked for viruses but the contents of an attachment may still contain software viruses, which could damage your computer system: you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

2 1

Thickness Repeatibility of a roi
by Campanella Martina 23 Mar '09

23 Mar '09

Hi, We would like to test the thickness repeatibility in a particular region of interest from 6 scans of the same subject in the same session and from the same subject in 6 different sessions to compute the overall statistics of them. We are trying to follow the protocol described in the wiki page about Thickness Repeatibility (http://surfer.nmr.mgh.harvard.edu/fswiki/ThicknessRepeatibility) but we're encountering some problems. We defined with tksurfer tools the label corresponding to the roi intended, afterwards we converted it in an annotation. The tutorial suggests the use of mris_thickness_diff but we have 6 scans, rather than only 2. So what can we do? Anyway we tried to find the thickness difference map using the first scan as the reference surface and the others as the input surfaces but, since we'd like to analyze the thickness deviation standard in a specific region, we used the option flag -annotation and the results seem wrong: we tried to use several annot files, specifing different label names but we found always the same error: total 0 vertices involved in thickness comparison. Any suggestions for our data? Thanks, Elisa Molinari, Martina Campanella

2 1

RE: [Freesurfer] merging rh and lh surfaces
by ppj＠netfilter.com.br 22 Mar '09

22 Mar '09

I think Bruce or Douglas can explain better. However, the reason for both hemispheres being separated is that you need to mantain an S2 topological space. Sent from my Nokia phone @E71 -----Original Message----- From: slew(a)nmr.mgh.harvard.edu Sent: 19/03/2009 19:07:48 Subject: [Freesurfer] merging rh and lh surfaces Hi, I would like to have a single surf file that composed of both right and left pial surf. Is there a way to do it in freesurfer? Best. seok -- Seok Lew, Ph.D. Athinoula A. Martinos Center for Biomedical Imaging Massachusetts General Hospital Building 149, 13th Street, Mailcode 149-2301 Charlestown, MA 02129-2060 USA e-mail slew(a)nmr.mgh.harvard.edu FAX 617 726 7422 _______________________________________________ Freesurfer mailing list Freesurfer(a)nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

5 5

Re: RE: [Freesurfer] Voxel size smaller than 1x1x1mm?
by "Falk Lüsebrink" 21 Mar '09

21 Mar '09

Hi Nick, yes I have been trying to use the -noaseg flag, but as recon-all did exit with an error I thought this process might be needed, so I haven't tried anything else. The error message I received was: #-------------------------------------------- #@# WM Segmentation Wed Mar 18 10:14:42 CET 2009 cp wm.mgz wm.seg.mgz mri_segment -keep brain.mgz wm.seg.mgz preserving editing changes in output volume... doing initial intensity segmentation... using local statistics to label ambiguous voxels... computing class statistics for intensity windows... WM (106.0): 106.9 +- 5.8 [80.0 --> 125.0] GM (69.0) : 66.3 +- 11.6 [30.0 --> 96.0] setting bottom of white matter range to 77.9 setting top of gray matter range to 89.4 doing initial intensity segmentation... using local statistics to label ambiguous voxels... using local geometry to label remaining ambiguous voxels... reclassifying voxels using Gaussian border classifier... removing voxels with positive offset direction... smoothing T1 volume with sigma = 0.250 removing 1-dimensional structures... thickening thin strands.... 20 segments, 4813 filled 10270 bright non-wm voxels segmented. 5589 diagonally connected voxels added... white matter segmentation took 2.4 minutes writing output to wm.seg.mgz... ERROR: mri_segment-MRIcheckVolDims: volume1 height=256 != volume2 height=320. Regards, Falk -------- Original-Nachricht -------- > Datum: Sat, 21 Mar 2009 10:08:24 +0100 > Von: "Falk Lüsebrink" <falk.luese(a)gmx.net> > An: falk.luese(a)gmx.net > Betreff: RE: [Freesurfer] Voxel size smaller than 1x1x1mm? > Falk, > > Probably you should try adding the -noaseg flag to recon-all, as the > aseg step relies on having 'conformed' volume data to register to the > atlas. The -noaseg flag skips creation and usage of the subcortical > atlas. > > Nick > > > On Wed, 2009-03-11 at 13:16 +0100, Falk Lüsebrink wrote: > > Hi again, > > > > so I did use the -cm flag to conform the voxels to its minimum size of > .8mm > > iso (which is also the native size of my data set) during the > mri_convert > > process and used the -noconform flag for the mri_normalize stage during > > autorecon1 to prevent further conformation to 1mm. > > > > But now recon-all exits with an error at stage 12 > > (http://surfer.nmr.mgh.harvard.edu/fswiki/recon-all) The error message > I > > got is as follows: > > #-------------------------------------------- > > #...@# Intensity Normalization2 Wed Mar 11 08:38:44 CET 2009 > > /home/falk/freesurfer/subjects/claus/mri > > > > mri_normalize -f home/falk/freesurfer/subjects/claus/tmp/control.dat > > -noconform -aseg aseg.mgz -mask brainmask.mgz norm.mgz brain.mgz > > > > using control points from file > > /home/falk/freesurfer/subjects/claus/tmp/control.dat... > > not interpolating and embedding volume to be 256^3... > > using segmentation for initial intensity normalization reading from > > norm.mgz... > > mri_normalize: aseg volume aseg.mgz must be conformed using MR volume > > brainmask.mgz to mask input volume... > > Linux xxxxxx 2.6.26-1-amd64 #1 SMP Mon Dec 15 17:25:36 UTC 2008 > > x86_64 GNU/Linux > > > > recon-all exited with ERRORS at Wed Mar 11 08:38:55 CET 2009 > > --------------------------------------------- > > > > Does anyone know what to do? > > > > Regards, > > Falk > > > > > > -----Original Message----- > > From: Bruce Fischl [mailto:fis...@nmr.mgh.harvard.edu] > > Sent: Friday, March 06, 2009 1:26 PM > > To: Falk Lüsebrink > > Cc: freesurfer(a)nmr.mgh.harvard.edu > > Subject: Re: [Freesurfer] Voxel size smaller than 1x1x1mm? > > > > Hi Falk, > > > > yes, you can use the "conform to min" (-cm I think) switch in recon-all. > > Can you cover the whole brain at .7mm iso? That's hard. And 7T is hard > > for whole brain morphometry because of dielectric effects. Good luck. > > > > Bruce > > On > > Fri, 6 Mar 2009, Falk Lüsebrink wrote: > > > > > Hi Freesurfers, > > > > > > I'm currently doing my bachelors thesis in medical engineering and I'm > > > trying to measure the cortical thickness of scans acquired by a 7T > MRT. > > > > > > A colleague of mine has used Freesurfer (v4.0.5) before and mentioned > that > > > high resolution scans with a voxel size smaller than 1x1x1mm (like > > > .7x.7x.7mm) are converted to 1x1x1mm during a processing stage of > > autorecon > > > - I think during the mri_convert process. > > > > > > Is it possible to keep the native size or does Freesurfer need to > change > > the > > > voxel size to 1mm in order to do the segmentation correctly? > > > > > > Regards, > > > Falk > > > > > > _______________________________________________ > > > Freesurfer mailing list > > > Freesurfer(a)nmr.mgh.harvard.edu > > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > > > > > > > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer(a)nmr.mgh.harvard.edu > > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > > > _______________________________________________ > Freesurfer mailing list > Freesurfer(a)nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > -- > Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: > http://www.gmx.net/de/go/multimessenger01 -- Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger01

1 0

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer March 2009