It depends a bit on whether they are nuisance regressors or not. If they
are not nuisance regs, then I would be very careful normalizing them
because it could make the regression coefficients hard to interpret when
comparing across subjects (you would probably just use the same scale
for all su
In a functional analysis, we have a set of external regressors (e.g. mean
luminance of stimuli, size of stimuli, etc.). Should we normalize values
before including them in the regressor files? If so, what is the
recommended way of normalization ... z-scoring?
Thanks,
Reza
_
I already used this argument during preprocessing.
According to the explanation here:
http://surfer.nmr.mgh.harvard.edu/fswiki/FsFastFunctionalConnectivityWalkthrough
in step-3, time-series data are sampled onto fsaverage then I am not sure
why 2nd-level analysis is showing me significant self conn
I would suggest you to check if preproc is done using the argument
-mni-305-2mm.
On Wed, May 11, 2016 at 12:48 PM, Sahil Bajaj wrote:
> Hi Dr. Greve,
>
> This works for me if I define seeds into subject-wise anatomical space.
>
> Now, location of seed regions across subjects might differ because
Hi Dr. Greve,
This works for me if I define seeds into subject-wise anatomical space.
Now, location of seed regions across subjects might differ because these
are defined in subject-wise anatomical space. Hence subject-wise
connectivity maps are generated from subject-wise seeds but these can be
You should map the ROI into the individual subject's anatomical space,
the specify the ROI when you configure the seed (eg, fcseed-config -seg
ROI.mgz -segid 1 ...)
On 05/10/2016 10:37 AM, Sahil Bajaj wrote:
> Hello all,
>
> I am interested in importing some seed regions which are in MNI space
Hello all,
I am interested in importing some seed regions which are in MNI space in
order to calculate functional connectivity maps. Looks like command
preproc-sess doesn't normalize the BOLD data and only the regions generated
by FreeSurfer LookUp table which have 'segid', can be used to calculat
I'm not sure what you mean by either of those options. can you explain?
On 10/22/2015 04:18 AM, geschwind2013 wrote:
> Hi, FS Experts,
>
> I’m using version 5.3 and have performed hippo-subfields analysis.
>
> To perform the statistics, shall I get the normalized volumes of each
> hippo-subfield
Hi, FS Experts,
I’m using version 5.3 and have performed hippo-subfields analysis.
To perform the statistics, shall I get the normalized volumes of each
hippo-subfields?
or just to do correction for ICV is enough ?
thanks..
Hyon
___
Freesurfer ma
Hi, FS Experts,
I’m using version 5.3 and have performed hippo-subfields analysis.
To perform the statistics, shall I get the normalized volumes of each
hippo-subfields?
or just to do correction for ICV is enough ?
thanks..
Hyon
___
Freesurfer
Hi Teo,
You might just need to adjust the contrast/brightness settings when
viewing the volume in freeview. If you move your mouse around do any
non-zero values appear in the bottom right voxel info panel?
The T1.mgz for bert should be 4.6mb, so you can see if what you have
matches that.
O
Hi Everyone,
I would like to normalize some images. But even when I test the -normalize
option with
recon-all -s burt/ -normalization
which from what I can see changes /burt/mri/nu.mgz (which came with the
freesurfer download) to /burt/mri/T1.mgz, T1.mgz is entirely black. Is
there a way to fi
Hi Maia
it's hard to say since I don't really know what steps worked in 5.3 or
why vs. 5.1. I guess you could try starting from after the normalization in
5.1 and seeing if they go through the rest of recon-all
Bruce
On Tue, 20 May 2014, Maia Pujara wrote:
Thanks for your response, Bruce.
Thanks for your response, Bruce. Since we would rather not re-run all 200+
subjects through version 5.3, would it be possible, for the sake of keeping
things consistent, to add any flags to recon-all using version 5.1, to be
able to include the remaining 35 subjects into our analysis?
On Fri, Apr
Hi Maia
Yes, it's possible. We've made a bunch of improvements
Bruce
> On May 15, 2014, at 1:13 PM, Maia Pujara wrote:
>
> To follow up, it makes sense that the noise might be an issue, but it seems
> like version 5.3 creates the surfaces just fine regardless of the noise, so
> I'm wondering i
To follow up, it makes sense that the noise might be an issue, but it seems
like version 5.3 creates the surfaces just fine regardless of the noise, so
I'm wondering if perhaps there has been a chance in preprocessing before
the end of the T1.mgz computation that solves the problem.
Would there ha
Hi Maia
it's hard to know from just these images, but I wonder whether the high
background noise level at the top of the image is messing up 5.1. You
might try cropping that out and seeing if things get better.
cheers
Bruce
On Fri, 25 Apr
2014, Maia Pujara wrote:
Hi all,
After running 200
FreeSurfers,
I have a couple of questions regarding normalization for brain/head size.
We are trying to analyze a pediatric population with subjects between age
4 and 16 with variable brain (and head) sizes. We are mainly interested in
measuring cortical thickness in these children,
but not sure
Hi Bruce,
The nu.mgz volume is has the most pronounced contrast between grey and
white matter. The orig.mgz volume blurs the distinction between white/grey
matter to a point where it is difficult to tell by visually inspecting it.
The grey matter in the orig.mgz volume has values as high as 147-15
Hi Gonzalo
it's hard to tell from just that image. What does it look like before
normalization (e.g. the nu.mgz or orig.mgz)? Is there gray/white contrast
there?
cheers
Bruce
On Wed, 8 May 2013, Gonzalo Iribarne wrote:
> Hello Freesurfers,
> There are some areas of grey and white matter that
It's clear now. Thank you very much.
And you're right about manual ICV, I have only two images for each patient
with the area of the biggest plan in the brain, commissural area.
Best regards,
Ana Arruda
___
Freesurfer mailing list
Freesurfer@nmr.mgh.ha
Hi Ana,
if you work on a study comparing groups or age effects, that analyzes
hippocampal volume, then you would usually normalize that volume. People have
different head sizes and therefore differently sized hippocampi. To remove
headsize differences researchers often normalize voumes by ICV.
Thanks a lot Martin.
I really get confused about normalization. I've already done the
comparisons and the values between methods, as you said, are different.
Although I haven't understood why, in this case, I don't have to normalize.
My advisor told me to normalize all the brain values from the pa
Hi Ana,
when comparing to manual labels or other segmentation methods you do not need
to normalize. But other methods and even manual segmentations may follow a
different protocol, so it can be expected that some method gives consistently
different (e.g. smaller) values.
ICV in FreeSurfer is c
Hello,
I'm working on hippocampus segmentations obtained by Freesurfer to compare
with manual and FSL segmentations. Do I have to normalize to make these
comparisons more accurate?
If so, what volume do I have to use? I've read about eTIV but I would like
to confirm if there is any value for norm
Not really, but you might want to account for total brain size or
(estimated) ICV.
doug
On 08/17/2012 02:13 PM, Alfredo Damasceno wrote:
> Dear freesurfers,
>
>
> Do I need to normalize volumes (MNI space or Tailarach) from
> aseg.stats and aparc.stats before entering data in another statistical
Dear freesurfers,
Do I need to normalize volumes (MNI space or Tailarach) from aseg.stats and
aparc.stats before entering data in another statistical programme?
Thanks
Alfredo
___
Freesurfer mailing list
Fre
Hi all,
Regarding the white matter parcellations in the wmparc.stats file, is it
necessary to normalize these values by eTIV or by another factor? Or
should this step be skipped as with the cortical thickness measures
(?h.aparc, etc.)?
Thanks!
-Jason
Hi Jeff,
I've seen this happen before with the N3 stuff where it reduces the image
intensities so much that you start losing dynamic range. Not sure what
causes it, but you could cp orig.mgz to nu.mgz and run from there
forward.
cheers,
Bruce
On Wed, 24 Sep 2008, Jeff Sadino wrote:
Hello
Hello all,
I am receiving an error when trying to run autorecon1. When trying to step
through it manually step-by-step, everything is fine up to -nuintensitycor.
After running that command and looking at nu.mgz in tkmedit, the picture starts
to fade out pretty bad above 120 intensity. When
Hi Glenn,
we have some estimated total intracranial volume (eTIV) in the new stuff
using something that Randy Buckner showed recently - the determinant of the
tal transform is predictive of TIV. Should be out soon. It's not clear
however that thickness needs to be corrected by it. The MNI fol
you mean by ICV? No, although in our new versions we compute ICV/TIV. Do
you mean volume labels or surface labels? Volume of volume labels can be
computed with mri_label_volume, surfaces can be given to
mris_anatomical_stats -l
cheers,
Bruce
On
Thu, 8 Dec 2005, Glenn Lawyer wrote:
Hi,
W
Hi,
We want to compare grey matter volumes from specific labels. Is the
output of mris_anatomical_stats normalized in any way? It doesn't
appear to be.
Thanks!
+glenn
--
I want to go farther, farther!
A. E. L.
> Glenn Lawyer <
> +352 061 967 244
Cc: freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] Normalization problem
Hi Martin,
we're looking into it - it's probably a bug in recon-all.
Bruce
On Mon, 21 Nov
2005, Martin Ystad wrote:
> I'm working on the latest developmental release for Rh. 9. and I'm
&g
Hi Martin,
we're looking into it - it's probably a bug in recon-all.
Bruce
On Mon, 21 Nov
2005, Martin Ystad wrote:
I'm working on the latest developmental release for Rh. 9. and I'm
having problems with the normalization procedures on my datasets.
My datasets have large intensity inhomogene
I'm working on the latest developmental release for Rh. 9. and I'm
having problems with the normalization procedures on my datasets.
My datasets have large intensity inhomogeneities due to their
acquisition with a surface coil (8-ch. GE).
To fix this problem I have resorted to the use of control p
I'm working on the latest developmental release for Rh. 9. and I'm
having problems with the normalization procedures on my datasets.
My datasets have large intensity inhomogeneities due to their
acquisition with a surface coil (8-ch. GE).
To fix this problem I have resorted to the use of control
Hi,
I am having trouble removing the pons with recon-all -stage1 -pons-xyz.
I've tried five pons coords, unsuccessfully.
When you move the mouse location in the sagittal, horizontal, and coronal
images,
the slice numbers change. When I get the talairach coords for the pons,
should the slice numbe
Hi,
For one subject's data, "process volume" didn't get the skull stripped.
The skull was there intact in "brain" volumes.
To correct it,
I used "expert preferences" and used "don't use watershed analyze", which
led to most of the skull removed in "brain" volumes.
But still there are pieces of th
You can tell recon-all not to touch the talairach.xfm file by adding
-notalairaach to the command-line
doug
On Thu, 6 Jan 2005, Joongnam Yang wrote:
Thanks Bruce,
How would I specify the new talairach.xfm in the run of recon-all -stage1.
If I copy the new file into the usual /mri/transforms directo
Thanks Bruce,
How would I specify the new talairach.xfm in the run of recon-all -stage1.
If I copy the new file into the usual /mri/transforms directory, it's
overwritten
during the recon-all process.
Comments are appreciated.
Nam.
>>> Bruce Fischl <[EMAIL PROTECTED]> 01/06/05 4:55 PM >>>
chec
check the talairach transform
On Thu, 6 Jan 2005, Joongnam Yang wrote:
Hi,
During recon-all -stage1 running, I got the following error.
...
MRInormalize: could not find any valid peaks
No such file or directory
mri_normalize: normalization failed
No such file or directory
ERROR: mri_normalize e
Hi,
During recon-all -stage1 running, I got the following error.
...
MRInormalize: could not find any valid peaks
No such file or directory
mri_normalize: normalization failed
No such file or directory
ERROR: mri_normalize exited with non-zero status
-
Any suggestion would be appreciated.
Na
43 matches
Mail list logo
