Dear Doug,
Thank you for the repy but may be couple of things. One, when I used the
above command it said segmentation volume is missing , so I added the
segmentation volume as in the command below and it worked and I also got
the out.dat but not as I wanted. With this command :
mri_segstats --se
Hi,
there is a bug in the scripts bedpostx_mgh and fsl_sub_mgh. Both use code that
do not conform to the standard shell syntax and work only with bash. This does
not work in system that do not link /bin/sh to /bin/bash (as is the case in
Ubuntu). Both scripts must use #!/bin/bash as the interpr
Hi Doug,
That doesn't seem to work either? Is there anything else I should try/ does
my fsgd file look alright?
Thanks for your help,
Jen
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Douglas N Greve
Sent: 24 January 2017 04:28
To: freesu
Dear Freesurfer Developers,
For my research I'm trying to run the hippocampal subfield segmentation
analysis however, I keep running into the following error. I tried to solve it
by myself however I had no luck ( I tried by removing the exec flag from the
files ).
I'm using the following sofft
Dear Freesurfer experts:
My project objective is to build 3D reconstruction of brain with
inserted electrodes. I want register the Brian volume of CT and MR via
freesurfer. The registration result looks great when I used th "mri_coreg"
instruction, but when I tried to use the information in
Dear freesurfer experts,
Is it possible to create Habenula as region of interest from fsaverage
(orig.mgz images) using tkmedit? I will be using following links to guide
me for creating Habenular region of interest. Hope my question is answered
since the next question is related to this please.
h
Hi Sampada
we don't automatically label habenula yet, but certainly you can draw it
using freeview. Was that your question?
cheers
Bruce
On Tue, 24 Jan 2017, Dr Sampada Sinha wrote:
Dear freesurfer experts,
Is it possible to create Habenula as region of interest from fsaverage
(orig.mgz im
Hi Matt
looks like you ahve a pretty big (almost 18K vertices) defect. It will
eventually fix it, but possibly not in the correct way. You can load the
lh.inflated.nofix in freeview and theoverlay lh.defect_labels to see
what's going on (this is the first defect). It may be that some of the
sk
Hi Matt
you only need to edit them if the lh.orig.nofix surface includes them.
Otherwise they are fine. The first defect will be near the back of the
brain
cheers
Bruce
On Tue, 24 Jan 2017, Hibert,
Matthew Louis wrote:
> Hi Bruce,
> Thanks, it does look like there is some skull included in t
Hi Bruce,
Thanks, it does look like there is some skull included in the brainmask.mgz and
wm.mgz in many locations (very noisy scan). I'll edit those out and give it
another try.
Matt
From: freesurfer-boun...@nmr.mgh.harvard.edu
[freesurfer-boun...@nmr.
Thanks Douglas. I am not sure that I understand. Will you please review the
commands below?
I believe I need to complete something like:
mri_surfcluster --in ${qdecoutputfolder}/cache.th40.abs.sig.masked.mgh
--fdr 0.05 --mask ${qdecoutputfolder}/mask.mgh --csd
/usr/local/freesurfer/average/mult-c
Hi Alex
cut it for what?
Bruce
On Tue, 24 Jan 2017, Alex Cohen wrote:
> So what resolution would make sense to use? the release notes say <1.0mm but
> would 0.9mm isovol really cut it?
> -Alex
>
>
> Alexander Li Cohen, M.D., Ph.D
So what resolution would make sense to use? the release notes say <1.0mm
but would 0.9mm isovol really cut it?
-Alex
Alexander Li Cohen, M.D., Ph.D.
E-mail: alexander.coh...@childrens.harvard.edu (Medical/Science Email)
E-mail: alex
yes, you can use mri_surf2surf to map the thicknesses to fsaverage space
(this is what recon-all -qcache does)
cheers
Bruce
On Tue, 24 Jan 2017, Onur Uğurlu wrote:
Dear Bruce,
Now I can get thickness values, vertex locations and the faces. But it seems
every subject has a different numb
Dear Bruce,
Now I can get thickness values, vertex locations and the faces. But it seems
every subject has a different number of vertices. In my case Subject1 has
107107 vertices, whereas Subject2 has 133093. Is it possible to normalized the
subjects into same space?
Thanks for the all help.
s
sorry, I still don't understand. Are you really trying to generate a
subcortical atlas? Or do you mean a subcortical segmentation? If the
latter, then 1mm is fine. The resolution you want really depends on what
you are trying to do
On Tue, 24 Jan 2017, Alex Cohen wrote:
> generating the subcor
generating the subcortical atlas:
recon-all now produces aseg.mgz (subcortical atlas) with Hi-Res data
(<1mm). The -hires flag is still necessary to include with recon-all when
hi-res data is input. Changes to mri_normalize, mri_em_register and
mri_watershed were made to support this feature.
--
It was more a question of:
Should the -hires switch be used if you're just interested in subcortical
parcellation, and your data happens to be 0.9mm, etc... or is this feature
aimed at ex-vivo data or similar data.
-Alex
Alexander
Dear FreeSurfer Developers,
I'm attempting to run recon-all on a test subject (through recon-all –all
–subjid test001 in Freesurfer 6). Recon-all hangs at the mri_ca_register
command throwing a Segfault error.
This is the corresponding part of the recon-all.log:
---
Thank you so much Andrew, I indeed looked into that. FLAIR volume used were
acquired in the same session, however not sure of different geometry. I had to
also fix the origins for FLAIRraw.mgz as their origins were outside the brain
(not sure if that would be contributing to different geometry,
Hi Florian
can you tar and gzip that subject dir and we will take a look?
cheers
Bruce
On Tue, 24
Jan 2017, Florian Krismer wrote:
Dear FreeSurfer Developers,
I'm attempting to run recon-all on a test subject (through recon-all –all
–subjid test001 in Freesurfer 6). Recon-all hangs at the
sure, it will preserve the 0.9mm which should help a bit if you have enough
SNR
On Tue, 24 Jan
2017, Alex Cohen wrote:
It was more a question of:Should the -hires switch be used if you're just
interested in subcortical parcellation, and your data happens to be 0.9mm,
etc... or is this featur
Hi there
I am wondering if the Freesurfer 6.0 version is more flexible with the
parameters used for ex-vivo scans other than what 5.3 required, as listed
here
https://surfer.nmr.mgh.harvard.edu/fswiki/ExVivo.
We acquired some ex-vivo scans at 0.4mm iso but not with those above
parameters.
many
Hi Trisanna
it is better about handling highres data, but the ex vivo recon stream is
still a work in progress
cheers
Bruce
On Tue, 24 Jan 2017, Trisanna Sprung-Much wrote:
Hi there
I am wondering if the Freesurfer 6.0 version is more flexible with the
parameters used for ex-vivo scans other
thanks for the quick response Bruce!
--
Ph.D. Candidate
McGill University
Integrated Program in Neuroscience
Psychology
On Tue, Jan 24, 2017 at 1:18 PM, Bruce Fischl
wrote:
> Hi Trisanna
>
> it is better about handling highres data, but the ex vivo recon stream is
> still a work in progress
>
Ah great! If the registration is accurate with bbr-init-spm, I suggest just
using that. I would only use the mri_robust_register technique if none of the
bbr initialization options produce a precise alignment. To overwrite the FLAIR
volumes, I'm pretty sure including the "-clean-FLAIR" flag in y
Hi Florian
how much RAM do you have in that machine? I think it is not enough
cheers
Bruce
On
Tue, 24 Jan 2017, Florian Krismer wrote:
Dear FreeSurfer Developers,
I'm attempting to run recon-all on a test subject (through recon-all –all
–subjid test001 in Freesurfer 6). Recon-all hangs
Hi list,I wonder whether the recon-all -all processing are running
correctly.I'm noting the follow linesThanks
Stefano
nsamples 3243
Quasinewton: input matrix
1.05460 -0.04529 -0.00589 -5.39976;
0.04715 1.21082 0.14343 -52.53261;
0.00057 -0.10515 0.95251 8.62344;
0.0 0.0
Perfect, shall go ahead with bbr-init-spm and will try running it with
"-clean-FLAIR" flag.
Thanks again!
Sneha
From: Hoopes, Andrew
Sent: Tuesday, January 24, 2017 2:15:31 PM
To: Sneha Pandya; Freesurfer support list
Subject: Re: [Freesurfer] mris_make_surfac
Hi Bruce,
it is a virtual machine with 8GB RAM dedicated to it. Any idea how much RAM
would be enough?
Thanks,
Florian
Am 24.01.17, 21:39 schrieb "Bruce Fischl"
:
Hi Florian
how much RAM do you have in that machine? I think it is not enough
cheers
Bruce
On
Hi Stefano
that's a warning burried in the depths of some open source quasi-newton
optimization code that doesn't seem to matter - it can be safely ignored I
believe
cheers
Bruce
On Tue, 24 Jan 2017,
std...@virgilio.it wrote:
Hi list,I wonder whether the recon-all -all processing are runn
I think that should be enough, but your data doesn't segfault for me. I
also noticed the following in your recon-all.log file:
usr/local/freesurfer/bin/recon-all
-all -gca RB_all_2017-01-23.gca -gca-skull RB_all_withskull_2017-01-23.gca
-subjid vco1573test_mpr
are you using your own version
Hi Bruce,
It’s a custom atlas with minor changes that was build with the
rebuild_gca_atlas.csh script.
What is puzzling to me is the fact that if I omit the align-after flag the
mri_ca_register command works perfectly fine.
Many thanks & best wishes,
Florian
Am 24.01.17, 21:48 schrieb "Bruce
Thanks!
>Messaggio originale
>Da: "Bruce Fischl"
>Data: 24-gen-2017 21.45
>A: , "Freesurfer support list"
>Ogg: Re: [Freesurfer] recon-all -all FS6.0
>
>Hi Stefano
>
>that's a warning burried in the depths of some open source quasi-newton
>optimization code that doesn't seem to matter -
Thanks for the tip, Peter. Someone has had this issue in past. These were
adapted from old FSL scripts and the goal is to phase them out eventually.
Right now trac-all has a new option to print out the commands to be
submitted to a cluster, so that the user can handle that herself. It would
be impr
Hi list,I'm performing hippo-subfields analysis on T1 having the followed
parameters:
data_type INT32dim1 170dim2 256dim3 256dim4
1datatype 8pixdim11.25pixdim21.00pixdim3
1.00pixdim41.00cal_max
Hi FLorian
I think you are not running the release, but a slightly earlier version
(from the info at the top of your recon-all.log).
Can you try running with the standard atlases and see if that works? And if
it does, upload your rebuilt atlases so I can try to replicate the crash?
cheers
B
hi freesufer folks, has your new version 6 run a successful cuda7.5 gpu on
ubuntu 16.04? i'm getting a core dump. jeff
here is the cmd:
recon-all -all -use-gpu -s ${filenames_arr[$i]}_freesurf -i
${sesnum_arr[$i]}/anat/${filenames_arr[$i]}.nii
first output gave:
Testing for CUDA device:
/home/
The -spm will work well, but it does require matlab (and spm). Note that
in version 6 we use mri_coreg which is our implementation of spmreg.
On 01/24/2017 02:15 PM, Hoopes, Andrew wrote:
>
> Ah great! If the registration is accurate with bbr-init-spm, I suggest
> just using that. I would only
Thank you Dough, yes I believe so, we have matlab and spm installed as well.
From: freesurfer-boun...@nmr.mgh.harvard.edu
on behalf of Douglas N Greve
Sent: Tuesday, January 24, 2017 5:00:19 PM
To: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] mris_m
Dear Martin,
thanks for the feedback. We had to edit cross due to the fact that
1. our dataset is involved also in cross-sectional studies
2. we are continuously acquiring timepoints but want to do some analyses with
data we already have
Therefore only editing the base was not an option for us.
Hi Antonin,
I see. The problem with the intersection would be that it could be zero in
spite of several edits that move around. Still would be better than nothing.
Feel free to test that (e.g. you can load that in matlab, intersect and then
apply to the wm from base to see if it helps). Let me
Hi FLorian
the problem is that you have cortical labels from the aparc+aseg in your
.gca file, and these exceed the max label we are expecting in a .gca atlas.
Is there a reason you need them and not just left and right cortical gray
matter labels? I don't think labeling folds will be terribly
Freesurfer devs,
Freeview is really nice to use. I have a few more suggestions:
It would be good to have keyboard shortcuts to change the brush size ala.
Tkmedit.
A clone feature ala tkmedit that would be needed to fill in brainmask or
brain.finalsurfs.manedit.mgz
When you are saving control po
44 matches
Mail list logo
