For #2, make_average_subject will create the .tif file in the output
directory.
doug
On 06/04/2012 09:06 AM, Bruce Fischl wrote:
Hi Christina,
1. Displaying data on your own subjects average can be less misleading
than on our's, particularly if you don't have many subjects. It shows
at
Hi Knut, you can copy the following file into
$FREESURFER_HOME/bin/mri_segstats
ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/mri_segstats.linux
(assuming you're using linux)
The run
recon-all -segstats -s subject
This will create a new aseg.stats file with total brain volume
They rely on the registration. What is failing? It is usually the
initialization that fails. If you're using FSL, you can try using the
SPM registration (though it's a little trickier to set up).
doug
On 06/04/2012 05:20 AM, Jörg Pfannmöller wrote:
Dear freesurfers,
I have a simple question
Hi everyone,
What is the best way to concatenate transforms (register.dat files, .lta files,
etc.)? I'm interested in registering functional data to a structural in
another session using a functional to in-session anatomical registration
combined with a in-session anatomical to out-of-session
Why not just register the functional to the desired structural directly?
That will work better than concatenation I think.
doug
On 06/04/2012 10:31 AM, Clark Fisher wrote:
Hi everyone,
What is the best way to concatenate transforms (register.dat files, .lta
files, etc.)? I'm interested in
Dear Freesurfers,
I had a question about interpreting the results of the Monte-Carlo
simulation. It is my understanding that the clusters are labeled based on the
value of the maximum significant vertex in a cluster. So, assuming that a
cluster kind of borders or extends over 2
That is correct and expected. The only constant is the vox2vox
transformation. Once you change the vox2ras (eg, scanner or TKReg), then
it must change the ras2ras transform in order to keep the vox2vox constant.
doug
On 06/02/2012 12:03 PM, Shay Ohayon wrote:
Dear Doug,
Following up on
Hi Hiroyuki, can you give us more information? Eg, you click on a given
vertex (col,row,slice), you find the Scanner RAS field in tkmedit, you
go to the same col,row,slice in freeview and it's RAS is x,y,z. There
are several different RAS coordinates, and you'll need to be clear as to
which
Hi Koushik, you are correct it will report the label that the max falls
into and so the entire cluster can fall across multiple labels. But I'm
not sure what you are asking. You're cluster effectively forms it's own
label, and, given that your cluster is somewhat unique, I'm not sure
what
OK, I put a new version of xhemireg there that will create the nu.mgz.
The newer version of the talairach do not need nu.mgz which is why I did
not have it in the xhemireg I gave you.
doug
On 06/02/2012 11:12 AM, Irene Altarelli wrote:
My command was:
xhemireg --s ac_copie
The terminal
Dear freesurfer community,
I tried to download rtview but it didn't work. I used the following URL:
ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/rtview
... as described here:
http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg21560.html
I need this function to
Hi Doug,
Attached is the qdec table file and associated levels files. It's strange
that qdec won't let me select more than one continuous variable. What's
even more strange is that I have been able to do that last month when I
originally looked at these data before doing manual edits and quality
OK, I've replicated that this is happening, but Nick will have to take a
look.
doug
On 06/04/2012 11:05 AM, Laura M. Tully wrote:
Hi Doug,
Attached is the qdec table file and associated levels files. It's
strange that qdec won't let me select more than one continuous
variable. What's
That link works for me, but I've attached rtview to this email just in case.
doug
On 06/04/2012 11:03 AM, Michael Bannert wrote:
Dear freesurfer community,
I tried to download rtview but it didn't work. I used the following
URL:
Dear all,
I have heard that the FWE correction sometimes could be demanding and kill
very subtle differences, overall for example in aging studies with normal
people and narrow-age intervals, or young people, where changes are
supposed to be not very big.
So, do you know when uncorrected
Thanks Doug for the response. Yes, I do understand that there is going to be
overlap. My question is, is there a way to identify all the overlapping
regions. Say, if the overlap is between prominent labels like Precentral and
superior frontal, even though the cluster label might say precentral,
Generally, reviews and editors will not let you publish without
correcting for multiple comparisons. You might get an exception with a
rare patient population or something like that. I don't understand your
second question.
doug
On 06/04/2012 11:22 AM, Daniel Ferreira wrote:
Dear all,
I
Nothing that I can think of, other than to just load it up in tksurfer
and see where the cluster falls.
doug
On 06/04/2012 11:18 AM, Govindarajan, Koushik Athreya wrote:
Thanks Doug for the response. Yes, I do understand that there is going to be
overlap. My question is, is there a way to
Dear Doug,
indeed the initialization fails. I was able to initiate bbregister manually,
giving a good registration result. Can you give me a hint how to
modify the registration and keeping the preproc-sess stream intact?
Cheers
pfannmoe
___
You can try running register-sess outside of preproc-sess, something like
register-sess -s session -fsd bold -per-run -init-spm
before you do that, you need to have spm5 and matlab, and change your
matlab/startup.m to add the spm5 path
path(path,'/usr/pubsw/common/spm/spm5');
doug
ps. Is this
Hi Kiley, that appears to be a problem with the MNI nu_correct program
(FSL unrelated). Do you change or re-install any of the MNI libraries?
Can you send the full recon-all.log file?
doug
On 06/02/2012 08:43 AM, Kiley Seymour wrote:
Dear Freesurfers,
I have previously had no trouble using
Hi Doug,
I do plan to manually align the functionals to the structural as a final step,
but our epi scans look different enough from our T1s that I want to use this as
a starting point. Also, combining transforms is something I've wanted to be
able to do a few times now as I explore different
You can use mri_matrix_multiply. Use -im to input a matrix, use -iim to
input a matrix that will be inverted.
doug
On 06/04/2012 12:13 PM, Clark Fisher wrote:
Hi Doug,
I do plan to manually align the functionals to the structural as a final
step, but our epi scans look different enough from
Hi, Douglas
Thank you very much for the updated one. I tested the new codes and the problem
is still there. The ID for the original version that generates a mask with one
missing ROI is:
$Id: mri_label2vol.c,v 1.38 2011/09/30 15:17:59 greve Exp $
I remember this is a version I downloaded from
Can you upload the subject to our file drop system?
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
On 06/04/2012 12:24 PM, Leon wrote:
Hi, Douglas
Thank you very much for the updated one. I tested the new codes and
the problem is still there. The ID for the original version that
so it seems that i ran make_average_volume which does have a --no-aseg flag
make_average_subject really just runs 2 other commands
make_average_surface and make_average_volume.
bruce the reason i did this is that mike was running a `super` average, i.e.
the average
subject of 2 different
FYI when the --no-aseg option is used with make_average_subject, it gets
passed to make_average_volume
doug
On 06/04/2012 01:56 PM, Gregory Kirk wrote:
so it seems that i ran make_average_volume which does have a --no-aseg flag
make_average_subject really just runs 2 other commands
Hello,
I'm running bedpostx_seychelles from the seychelles prompt (via trac-all
-bedp -c $SUBJECTSDIR/[dmrircfile]) with vers. v5.1.0-20110514. The
processes are generated and run correctly, as far as I can see, until the
post-processing job. I receive and email stating:
PBS Job Id:
Hi Sean,
You should look at the Exit_status. If it is '0', then your job ran
successfully. If it is '1' or something else, then there is an error. So
it looks like your job ran correctly.
Priti
Hello,
I'm running bedpostx_seychelles from the seychelles prompt (via trac-all
-bedp -c
Hi, Douglas
I have uploaded the file to the system.
Also, my new results showed that when ic3.tri (instead of ic2.tri) is used as
the input, one ROI at the similar location is also missing. Hope it will help
you diagnose.
Thanks
Leon
From: Douglas N
Hi Priti,
Thanks, but would seem that despite the 0 status, something didn't run
correctly. trac-all -path cannot find the file merged_ph1samples.nii.gz.
Actually none of expected bedpostx files, other than nodif_brain_mask have
been created in the dmri.bedpostx folder.
-Sean
Hi Sean,
You
Do not run your jobs on seychelles. Most nodes are still CentOS4 and I
think a lot of the tools are simply not going to run on CentOS4 anymore.
Run them on launchpad and if you still get errors then we need to investigate.
-- Paul Raines (http://help.nmr.mgh.harvard.edu)
On Mon, 4 Jun 2012
Hi Sean,
Are you sourcing Freesurfer stable 5.1 in seychelles? trac-all -bedp is
supposed to submit multiple jobs (one job for each slice, in addition to
bedpostx_pre and bedpostx_post).. When you ran trac-all -bedp did it
submit all of those jobs?
Priti
Hi Priti,
Thanks, but would seem that
1. How do I register subjects' thickness maps to a common space to be able to
do vertex wise comparison? And can any brain map be used as the common space?
2. If I have a hand-drawn ROI on subject's original T1, is it possible to
resample the volume ROI to the surface (so I can extract
On 06/04/2012 03:52 PM, Jiang, Zhiguo wrote:
1.How do I register subjects’ thickness maps to a common space to be
able to do vertex wise comparison? And can any brain map be used as
the common space?
This is the same as doing a group analysis. Basically you run
mris_preproc, mris_fwhm to
OK, I have the data. Can you send me the exact command line I need to
run to replicate your results? No shell variables or wild cards ...
On 06/04/2012 02:44 PM, Leon wrote:
Hi, Douglas
I have uploaded the file to the system.
Also, my new results showed that when ic3.tri (instead of
Hi, FreeSurfer experts
I have one annotation file with many ROIs and one label file that denotes
medial wall. Now I want to remove the ROIs in the annotation file that is
covered by the medial wall. Is there any way that I can realize that in the
surface space?
Thank you!
Leon
so what i should be able to do is run make_average_surface and
make_average_volume separately with a --no-aseg flag on
make_average_volume.
On 6/4/12, Gregory Kirk gk...@wisc.edu wrote:
so it seems that i ran make_average_volume which does have a --no-aseg flag
make_average_subject really just
you should be able to run make_average_subject with --no-aseg
On 06/04/2012 06:25 PM, mdkrue...@uwalumni.com wrote:
so what i should be able to do is run make_average_surface and
make_average_volume separately with a --no-aseg flag on
make_average_volume.
On 6/4/12, Gregory
doug-
When i try to do this it still brings up an error with the
make_average_subject regarding the --no-aseg as a subject to be
averaged. I went ahead and ran make_average_surface and
make_average_volume with a --no-aseg flag. This should recreate the
process of the make_average_subject correct?
yes, that should work
On 6/4/12 6:53 PM, mdkrue...@uwalumni.com wrote:
doug-
When i try to do this it still brings up an error with the
make_average_subject regarding the --no-aseg as a subject to be
averaged. I went ahead and ran make_average_surface and
make_average_volume with a
Thanks Doug - Nick, do you have any idea what might be happening here? Why
might qdec stop allowing me to select more than one continuous variable?
Thanks!
Laura.
Message: 24
Date: Mon, 04 Jun 2012 11:19:57 -0400
From: Douglas N Greve gr...@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] qdec
Hello,
We would like to calculate a Total Brain Volume based on data in aseg.stat
generated from version 5.1.0. Can I use intra crainal volume as estimated for
total brain volume? Or should I just add total grey volume with subcortical
grey volume with total white matter volume?
Thanks,
Dear freesurfers,
I have a simple question concerning the freesurfer 5.1 FSFAST preprocessing. In
the preprocessing the functional images are registered to the anatomical ones
using bbregister. This is the 3rd in a series of 8 preprocessing steps. If the
preproc-sess registration fails it is
Hi all -
Having a bit of an issue with trying to run a make_average with the flag
--no-aseg. When written
make_average_subjects --subject subject1 subject2 --no-aseg --out
superaverage
the propt runs as if --no-aseg and --out superaverage are subjects i wish
to average in with subjects1 and
Hi,
many thanks for your help!
Unfortunately, I still have some questions:
1) Why would someone only want to display the results on a cohort-specific
average, but not use that average for calculation?
2) So if I want to use my cohort-specific average for calcuation of the
cortical thickness
THAT was the problem. Thanks so much Bruce.
I owe you one.
-David
On Sunday, June 3, 2012, Bruce Fischl fis...@nmr.mgh.harvard.edu wrote:
Do you change the vertex indices together 1-based in matlab? I don't see
it...
Bruce
On Jun 3, 2012, at 9:06 PM, David Groppe
Hi Michael
a couple of typos. It should be make_average_subject (note no s), and
it should be --subjects (note the ending s). I also don't see a flag
named --no-aseg, which I think is why it's being included as a subject.
What are you trying to do?
cheers
Bruce
On Mon, 4 Jun 2012, michael
Hi Christina,
1. Displaying data on your own subjects average can be less misleading than
on our's, particularly if you don't have many subjects. It shows at least
one measure of the true variability/accuracy of the registration of your
subjects.
2. To make a study-specific target for the
sure. Happy to help
Bruce
On Mon, 4 Jun 2012, David Groppe wrote:
THAT was the problem. Thanks so much Bruce.
I owe you one.
-David
On Sunday, June 3, 2012, Bruce Fischl fis...@nmr.mgh.harvard.edu wrote:
Do you change the vertex indices together 1-based in matlab? I don't see
it...
50 matches
Mail list logo
