The Department of Psychiatry at the University of California, San Diego is
pleased to invite applications for a postdoctoral research scholar position in
neuroimaging and behavior genetics under the direction of Drs. William Kremen
(Psychiatry) and Anders Dale (Radiology and Neuroscience). The
Postdoc Positions for the Adolescent Brain Cognitive Development (ABCD) study
University of California, San Diego
Several funded postdoctoral positions are available in image analysis method
development, including automated cortical- and subcortical segmentation,
advanced diffusion tracto
FYI, in version 5.1.0, FreeSurferColorLUT.txt has a space (" ") instead of an
empty line at line 1189. This causes the matlab funciton read_fscolorlut to
throw an error.
An easy fix is to change this line:
if(~isempty(tline) & tline(1) ~= '#') break; end
to this:
if(~isempty(deblank(t
I am trying to use mri_surf2surf to resample surface data in mgh format
from ico4 to ico3. There is an apparent bug, however, in that the output
file has the same dimensions as the input file despite srcicoorder = 4
and trgicoorder = 3. I am using ico for both srcsubject and trgsubject.
This
My suggested work-around in my earlier email does not actually work. It outputs
as ico7 even though my "average" subject is ico3.
From: dhagle...@hotmail.com
To: freesurfer@nmr.mgh.harvard.edu
Date: Thu, 17 Nov 2011 11:52:22 -0800
Subject: [Freesurfer] mri_surf2surf ico to ico
I am tryi
%
% dipole_info=ts_read_dip_file(dip_file)
%
% read freesurfer dip file for the location and orientation of the dipoles
% returns dipole_info containing x,y,z location and nx,ny,nz normal vector
%
% created:05/07/05 by Mingxiong Huang as read_dipdec.m
% early mod: 11/27/05 by Don Hagler as
...@lanl.gov; freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] write dipoles function
thanks Don
are you guys still using this function?
Bruce
On Wed, 25 Jan 2012, Don Hagler
wrote:
> Hi Bruce and Amy,
>
> I think that menu command actually outputs a binary file, which I would
Hi Doug,
What does the new version of mris_preproc do with area?
Don
Date: Mon, 27 Feb 2012 20:54:43 -0500
From: gr...@nmr.mgh.harvard.edu
To: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] In Qdec, what does the 'area' mean within 'Measure'
menu in 'Design' tab
matlab's dicomread can handle 8 bit images, so if you have matlab, one could
write a script to convert your dicoms to mgh format (using save_mgh).
> Date: Fri, 30 Jan 2009 16:06:52 -0500
> From: gr...@nmr.mgh.harvard.edu
> To: senu...@fas.harvard.edu
> Subject: Re: [Freesurfer] difficulty readin
It seems to me that vertex areas in the native subject space are not
meaningful. The orig surfaces should be uniformly tesselated; however,
topology correction and final surface finding should introduce complicated and
probably uninteresting variation.
But, if a subject's surfaces are resampl
You could extend the analysis described on that wiki page by calculating tstats
and pvalues (in addition to mean and stdev). Then you have a paired t-test,
whereas mri_glmfit with qdec would be doing an unpaired t-test. I think
preproc has an option to calculate within subject differences for
determine the exact size at
which clusters becomes significant given a specified significance on the
surface data.
Then i found there is a subject talking about this:
https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2007-August/005855.html
In that talk,Don Hagler mentioned Keith Worsley
-value (say,0.05) to compare this
size with the corrected one, to see how difference they are.
function [cluster_threshold,peak_threshold] =
fs_calc_cluster_thresh(nverts,area,fwhm,df,alpha,pval);
Thanks, regards,
Jidan
On Fri, Apr 3, 2009 at 11:49 AM, Don Hagler wrote:
Assuming Wors
I think you should average your polar angle and eccentricity maps across
session and then calculate fieldsign from that. You do a complex average (a
separate average for real and imaginary components).
By the way, the sign of the fieldsign measure should have a fixed meaning; the
orientation
, wiki page, etc. which explains some of
these details--if you happen to know of one, I would appreciate the reference.
Thanks,
Jeff
On Fri, May 8, 2009 at 3:32 PM, Don Hagler wrote:
I think you should average your polar angle and eccentricity maps across
session and then calculate fiel
rect?
Jeff
On Mon, May 11, 2009 at 2:45 PM, Don Hagler wrote:
1. Fieldsign is supposed to be -1 or 1, without values in between. It is the
sign of the cross product. I don't see why you would want to average the
fieldsign. You will get the best estimate of the fieldsign by calcu
When we run v430 on our cluster, running rocks redhat 4.3, we get the error
message below when mris_volmask is run. From a google search, I see that
someone had the same error message back in Oct 2007. Was a solution found for
this? It runs fine on our desktop machines as well as when we log
> >
> > In my experience the mri_volmask runs fine for quite a while and then
> > crashes.
> > >From the debug messages you show it would thus could be that the process
> > >may
> > hit some limit of the threads which could be allocated to a process.
>
urfer] vtkMultiThreader error
Pretty much what I was saying the other day. It looks like maybe you
can disable this step if you don't want this output, though?
- Forwarded message from dhagle...@hotmail.com -
Date: Mon, 8 Jun 2009 11:19:43 -0700
From: Don Hagler
Reply-To:
%
% Note: A FreeSurfer register.dat file looks like this
% (minus leading spaces and # comments):
% bert # subject name
% 1.875000 # in-plane voxel size
% 2.50 # slice thickness
% 1.00 # ?
% 9.969735e-01 -7.713004
Postdoctoral Research Scholar Position in Behavior Genetics and
Neuroimaging
Departments
of Psychiatry and Radiology, University of California, San Diego
The Departments
of Psychiatry and Radiology at the University of California, San Diego are
pleased to invite applications for a postdocto
Bruce recently wrote that you could run mris_expand like this:
mris_expand -t lh.white 0.5 lh.graymid
I found that didn't work, but this did:
mris_expand -thickness lh.white 0.5 lh.graymid
I am using version 4.4.
By the way, is there a way to increase the resolution of the surfaces before
cre
I recently started getting this error message when running tkmedit:
GLUT: Fatal Error in tkmedit.bin: failed to create OpenGL rendering context.
This is with v4.4 on a machine running RHE5. Recent updates to RHE5 may be to
blame.
I get this error when running v3.0.5:
freeglut
(/space/monkeys/
I was able to resolve my issue related to OpenGL causing tksurfer and tkmedit
to have problems, and the solution may be helpful to others.
The problem was that the kernel had been updated (e.g. yum update), causing
problems for the nvidia display driver. The solution was to uninstall the
nvid
Postdoctoral Research Scholar Position, Multimodal Imaging
Laboratory, University of California, San Diego
The
Multimodal Imaging Laboratory (MMIL) at the University of California, San Diego
invites applications for multiple postdoctoral research scholar positions under
the directi
Who says FDR is more conservative? It all depends on the thresholds you choose.
Note that the FDR threshold is different from the cluster-based p
threshold, and it isn't necessarily appropriate to use the same for
both.
The cluster-based p threshold controls the overall probability that you wil
, not GRF, so we don't have to worry about the GRF assumptions
> breaking down at higher p-values.
>
> doug
>
> Don Hagler wrote:
> > Who says FDR is more conservative? It all depends on the thresholds
> > you choose.
> >
> > Note that the FDR thres
In the stable3 version of mri_vol2surf, there is an option listed
"projfac-avg". I can't tell if it is working or not. In the output
messages, it says that ProjFrac = 0.5 (or whatever value is given with
-projfrac) but there is no mention of the projfrac-avg min and max. Is the
program doing
What's your cmd line? What's the output?
I've pasted the script I used and the output below. You can see that there
is no mention of projfract-avg in the output.
By the way, it seems that it is ignoring the voxel resolution in the
registration file and instead using the (erroneous) resoluti
Running recon-all with the v3.0.2 release, I have noticed recently that the
process will run into seg faults at various stages (e.g. topology
correction, sphere reg, or cortical parcellation). This seems somewhat
random, but judging from some other posts, and the fact that it happenened
when I
s)
% faces: vertex numbers for each face (3 corners)
% coords: x,y,z coords for each vertex
% nbrs: vertex numbers of neighbors for each vertex
%
% created: 03/02/06 Don Hagler
% last modified: 05/09/06 Don Hagler
%
% code for reading surfaces taken from Darren Weber's freesur
= fd3(fid)
% read a 3 byte integer out of a file
b1 = fread(fid, 1, 'uchar') ;
b2 = fread(fid, 1, 'uchar') ;
b3 = fread(fid, 1, 'uchar') ;
retval = bitshift(b1, 16) + bitshift(b2,8) + b3 ;
%
From: "R. Chen" <[EMAIL PROTECTE
In older versions of tksurfer, it was possible to set a renderoffscreen flag
(in a tcl script) that would make it possible to render images in the
background. I find this very valuable for rendering large batches of stats.
I would like to store surface-painted timeseries data in mgh format and
If there is interest, I could post my shell script for generating a tcl
script and running tksurfer (with an option to write to rgb). I have
another one that will render all w files in a particular directory (as
rgb's) and another that will render all frames from an mgh (painted surface)
file
Here you go. No guarantees of usability or support.
From: Chacko Cherian <[EMAIL PROTECTED]>
To: Don Hagler <[EMAIL PROTECTED]>, [EMAIL PROTECTED],
[EMAIL PROTECTED]
CC: [EMAIL PROTECTED], freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] printing out from tksurfer
I am getting an error running mri_normalize (as part of recon-all) on a
particular subject's volume.
Here is the command from recon-all:
mri_normalize -g 1 nu.mgz T1.mgz
Here is the output:
WM removal version
using max gradient = 1.000
reading from nu.mgz...
normalizing image...
talairach trans
I have a subject that has gotten stuck on the CA Normalize step. I have
tried running it several times and it always gets stuck in the same place,
running away for days if I let it, using 99.9% of the CPU. We are running
RH4-x86_64-R302.
Is this something that has been observed / dealt with
Hi,
I've found that the number of voxels for a given aseg ROI in aseg.stats does
not match exactly the number of voxels when I load the aseg.mgz in matlab
and find the voxels of the corresponding ROI code. e.g.
vol=load_mgh('aseg.mgz'); length(find(vol==17)).
Is there a reason for this diff
I was trying to warp a volume in atlas space (i.e. an ROI) to single subject
space using mri_convert -ait ... (version = stable3) and ran into trouble.
I found that mri_convert will quit if the data type is int (type 1) and will
give strange results if data type is float (type 3). I was able t
To get your coordinates into freesurfer space, you need to register them
with the T1 image. You could in theory do this with tkmedit, but in my
experience, the headpoints interface is broken.
I do the registration (manually) with a custom matlab program, but it
requires that you have generate
Just a comment: Surface area on a vertex-wise basis doesn't seem very
informative. Before topology fixing, the surface area should be about the
same for all vertices (differences should be more or less random). Topology
fixing will add or remove vertices, so the surface area in a "fixed" regi
Can someone please explain what transformation matrix is contained in
talairach.lta?
Is this the registration matrix from atlas to subject or subject to atlas?
Also, is this matrix for rigid body registration only, or does it have the
affine registration?
gt;
To: Don Hagler <[EMAIL PROTECTED]>
CC: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] talairach.lta
Date: Mon, 11 Jun 2007 23:44:20 -0400 (EDT)
Hi Don,
I believe it is from subject ras to talairach ras, and it is definitely
affine (not just rigid). The .lta format is generic en
In recon-all, it says you are supposed to use -autorecon2-cp if you use
control points. So I wonder why earlier normalization steps also use
control points, such as Norm1 and CANormalize. What effect do the control
points have in those steps?
_
So then for redoing a recon with control points, it would be beneficial to
start again at the mri_normalize step?
From: Nick Schmansky <[EMAIL PROTECTED]>
To: Bruce Fischl <[EMAIL PROTECTED]>
CC: Freesurfer Mailing List , Don
Hagler <[EMAIL PROTECTED]>
Subjec
Hi Marie,
I think I had a similar problem with my script. I believe the problem is
with the two
lines of chars (the creation string). For whatever reason, I think that it
worked better
if the second line was empty.
This is the relevant section from my script:
fs_fwrite3(fid,TRIANGLE_FILE_M
rea run
% fs_load_subj (or fs_read_surf) and fs_calc_triarea
%
% NOTE: this program uses Worsley's RFT method for t-stats
%
% NOTE: this code is provided with no warranty whatsoever,
% no gaurantee of usefullness, and no offer of technical support.
%
% created: 07/06/07 Don Hagler
% last modifi
There is a renderoffscreen flag that can be used from a tcl script.
Unfortunately, this feature was broken when a particular section of the
tksurfer code was commented out. The problem was that the tclscriptflag
used to determine whether to open a window was not getting set to true.
Look for
I processed some brains using version 4 and it seems that it took
substantially longer than typical for version 3. It took about 3 days for
one subject with version 4, compared to 1.5 days for the same subject with
version 3.0.5. Is this on par with what others have experienced?
At least for
I see that mris_make_surfaces is normally used to generate thickness and
curvature. I would like to generate the ?h.thickness and ?h.curv files
without running mris_make_surfaces (I already have white and pial surfaces).
Can mris_thickness and mris_curvature (or other programs) give the same
there any reason to think that mris_make_surfaces and mris_thickness
would have differences in how the thickness measurements are done? I'm
doing this with version 3.0.2 at present.
Don
From: Bruce Fischl <[EMAIL PROTECTED]>
To: Don Hagler <[EMAIL PROTECTED]&g
I looked at the code and found another option for mris_thickness: -N for
neighborhood size.
For mris_make_surfaces, the default is 20, but for mris_thickness, the
default is 2. So setting it to 20 gave the correct result.
From: "Don Hagler" <[EMAIL PROTECTED]>
To: [EMA
Are you sure that it is saving t-values in F.mgh and not F-stats?
From: Glenn Lawyer <[EMAIL PROTECTED]>
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] t-value saved values
Date: Thu, 20 Sep 2007 15:00:44 +0200
Hi,
I asked about this last week, and had a nice discussion with Doug, b
Every time point not specified as a condition would be counted as baseline,
right? So does that mean then, that you have a different baseline for your
3 and 5 condition analyses?
From: "Cameron Ellis" <[EMAIL PROTECTED]>
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] activation pat
If you have problems writing out mgh files with 14x1x1x500 volumes, maybe
you need more memory (>=2GB?). Or a 64 bit machine? Or you could downsample
your time courses if you haven't already.
I do use matlab to convert decimated source time course files (stc) to mgh.
Before writing out
Don, have you tried using mri_surf2surf with --mapmethod nnf ? That
will prevent the interpolation and map each ico4 to the nearest ico7.
doug
Don Hagler wrote:
If you have problems writing out mgh files with 14x1x1x500 volumes,
maybe you need more memory (>=2GB?). Or a 64 bit machi
It is easier to find the neighborhood for a given number of steps away from the
vertex of interest. One easy way is to set the value of a single vertex,
smooth for a particular number of iterations, and then select the non-zero
vertices. If you smooth 5 times, you will get 5 steps away. The
> surfaces), as the smoothing seems to introduce entirely new foci of
> activity, particularly around the insula. Therefore, I am trying to find
> a way to have a set of common source points so that smoothing can be
> avoided until the final step to display the data.
>
> Thanks in advan
matlab's reducepatch
Date: Thu, 10 Apr 2008 10:39:37 -0700
From: [EMAIL PROTECTED]
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] mri_watershed -> BEM surfaces problem
P.S. Also, I was wondering if there is a utility to reduce the number
of vertices of a given surface (i.e. to make a le
You can see that it is running matlab just before the error. So it must be
using the matlab load_mgh function, which does convert from mgz to mgh and then
deletes the tmp file when it is done. Maybe you are setting rm as an alias for
'rm -i' in your .cshrc.
> Subject: Re: [Freesurfer] LGI
External Email - Use Caution
I have recently started using FreeSurfer 6.0 after previously working with 5.3
for several years. I found that the lh.aparc.a2009s.stats files use different
ROI names (e.g., G&S_frontomargin) than they did in FS5.3
(G_and_S_frontomargin). However, th
Postdoctoral Research Scholar Position, Multimodal Imaging
Laboratory, University of California, San Diego
The Multimodal
Imaging Laboratory (MMIL) at the University of California, San Diego invites
applications for a postdoctoral research scholar position under the direction
of Drs. Donald
There is no version of freesurfer being distributed by Anders Dale at UCSD.
Marty Sereno does has a version of freesurfer branched from version 0.9. His
includes a gui called csurf that can launch the retinotopy analyses (among
other things) for a single subject. In general, it is somewhat d
Postdoctoral Research Scholar Position, Multimodal Imaging
Laboratory, University of California, San Diego
The Multimodal
Imaging Laboratory (MMIL) at the University of California, San Diego invites
applications for a postdoctoral research scholar position under the direction
of Drs. Donald
The attached file can by copied over the existing talairach.xfm (you might want
to make a backup copy).
To clarify for others reading this, you would only do this when you want to run
make_average_subject on a single subject so as to resample a subject's surface
to ico (useful for quantifying
We had exactly that problem on machines running RedHat 5, but it was solved by
reinstalling with the latest CentOS.
From: kc...@bu.edu
Date: Wed, 4 Aug 2010 15:44:52 -0400
To: fis...@nmr.mgh.harvard.edu
CC: freesurfer@nmr.mgh.harvard.edu; ni...@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] keyb
In my recent experience, no. There were problems with entering text in
tksurfer/tkmedit when running on the pre-installed RHEL that went away when
CentOS was installed later.
> Date: Mon, 9 Aug 2010 15:42:57 -0400
> From: rai...@nmr.mgh.harvard.edu
> To: dah...@nmr.mgh.harvard.edu
> CC: free
Is the 2.7GB RAM estimate new for v5? For v3 and v4, 2GB seems to have been
sufficient (using a hard limit on a cluster). We haven't switched to v5 yet
(waiting for 5.1 or 5.2), so would like to know what to expect when we do.
> From: ni...@nmr.mgh.harvard.edu
> To: daniel.geis...@gmail.com
>
The area maps do contain the actual surface area of each vertex with units of
mm^2. The problem is that the area at a given vertex depends on how many
vertices a given patch of
cortex was divided into, which is initially uniform but then varies
arbitrarily due
to topology correction, rend
Using version 4.5.0, I get this error message with mri_vol2surf:
INFO: Source volume detected as selxavg format
number of frames is incorrect (0,1)
This error does not happen with version 3.0.5, and it only happens for input
volumes with particular file names.
This fails:
mri_vol2surf \
--s
e is not used anymore, so I'll
> have mri_vol2surf stop trying to load it. For now, just name your .dat
> files differently than your other files.
>
> doug
>
> Don Hagler wrote:
> > Using version 4.5.0, I get this error message with mri_vol2surf:
> > INFO:
Hi,
In response to the message below, I have been getting the exact
same error when running tksurfer on a particular workstation.
tksurfer does display correctly, but the drawing is very slow, as if
running remotely (or not using the video card). This happens with versions
3.0.5 and 4.5.0.
I tr
A quick update on my earlier posting: After disabling the "Composite"
extension, I no longer
get the UBB warning in the Xorg.0.log, but I still get the error and slow
drawing with tksurfer.
From: dhagle...@hotmail.com
To: freesurfer@nmr.mgh.harvard.edu
Date: Tue, 15 Mar 2011 12:07:56 -0700
Sub
For retinotopy, I find it convenient to use the sphere surface. It does not
require a cut down the middle of V1 and it is flat enough to easily draw ROIs.
Also, in my experience, the patch flattening algorithm often produces poor
results (e.g. large distortions, twisting).
Date: Thu, 28 Apr
Hi Anders,
I agree with the points Jon made. 8 Hz flicker is also what I use. I haven't
tried lower frequencies, but
I think you would definitely have reduced SNR if you did. I also use rather
thin wedges and rings and smoothly vary their position.
I would add that a longer cycle duration
When I try to do a full flattening of a surface I often get areas with
extreme squishing, looking like wrinkles. Usually if I redo the cuts in
slightly different ways (e.g. adjust the spacing between cuts, extend the
cuts farther onto the lateral surface, etc) I can get the flattened surfac
Can anyone supply an example head points file for use with tkmedit?
___
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
When I run recon-all with the -autorecon2-wm flag, my white matter edits get
overwritten, presumably by aseg, even if I use the -keepwmedits flag. Is
that the correct behavior or is this a bug?
If instead of -autorecon2-wm I use -stage2, then the wm is untouched (as
desired). For some reason
Here is the version info for recon-all:
$Id: recon-all-nmr,v 1.107 2005/09/12 17:36:59 greve Exp $
From: Doug Greve <[EMAIL PROTECTED]>
To: Don Hagler <[EMAIL PROTECTED]>
CC: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] keeping wm edits
Date: Mon, 07 Nov 2005 16:08:02
I had a similar problem running Linux RedHat E4. It seems that fsl and
freesurfer have incompatible versions of tcltktix. One solution is to
always run freesurfer and fsl from different terminal windows.
Peter Soros wrote:
Wed, 02 Nov 2005 15:19:19 -0800
Hi All,
I have installed Freesurfe
I forgot to mention that your .cshrc or .bashrc should not automatically
initialize freesurfer or fsl. You would want to set up an alias to do that
selectively per terminal window.
Don Hagler wrote:
I had a similar problem running Linux RedHat E4. It seems that fsl and
freesurfer have
The problem does occur when fsl is initialized after freesurfer, but it may
be possible to comment out certain lines from the freesurfer setup script.
I haven't tried that.
From: Stephen Smith <[EMAIL PROTECTED]>
To: Don Hagler <[EMAIL PROTECTED]>
CC: freesurfer@nmr.mgh.ha
I want to average multiple structural scans for input into the surface
recon. Two are axial and one is sagittal. (I plan to correct for the
gradient inhomogeneities in the sagittal images, so that shouldn't be an
issue.)
Anyway, the problem comes when recon-all does "motion correction" to al
Sorry for being overly dramatic. recon-all just quits with error messages:
ERROR: TRs are inconsistent 8.012 7.996
ERROR: mri_motion_correct exited with non-zero status
From: Doug Greve <[EMAIL PROTECTED]>
To: Don Hagler <[EMAIL PROTECTED]>
CC: freesurfer@nmr.mgh.harvard.edu
I tried to run fieldsign calculation using surfaces created with the new
recon-all but the results look wrong, with some vertices that have a
salt-and-peppery look to them. The code I'm using is ripped out of an older
version of tksurfer and works fine on old surfaces.
It seems that there is
Hi guys,
There seems to be a minor bug in mri_surf2surf when trg_type = w. If
trgsurfval is given as a relative path name (e.g. no directory sepcified
with intention to create in current working directory), the output file name
can be prefixed by" lh." or "rh." and it can be put in the subje
I should have mentioned that a simple workaround is to precede the
trgsurfval with "./" to explicitly specify that it should be created in the
current working directory.
From: "Don Hagler" <[EMAIL PROTECTED]>
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfe
I tried painting to a surface created with the new recon-all (using paint or
mri_vol2surf (version 1.22 2005/07/05)), and it seems that the surrfaces are
not being read correctly (the data in the w file has the speckled appearance
you get when the vertex numbers are wrong).
I posted a message
oops, my dumb mistake. I forgot to change the subjet name in the
register.dat.
From: "Don Hagler" <[EMAIL PROTECTED]>
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] paint, mri_vol2surf, paint-sess
Date: Thu, 23 Feb 2006 14:26:14 -0800
I tried painting to a surface
89 matches
Mail list logo
