Hi Jon,
this is for the cross sectional stream, so it is the same as what you
did when you processed your data independently. You point recon all to
some input file and it generates the subject/mri/orig/001.mgz etc . You
can also do this in two steps, the -i to import your data and then the
-
OK,
so I should plug the nifti's as opposed to the freesurfer generated mgzs
jon
On Jul 1, 2013, at 11:35 AM, Martin Reuter wrote:
> Hi Jon
>
> That is one way of calling the regular recon-all stream. You point it to any
> input file with the -i flag. Here dcm could be a dicom image, but can b
Hi Jon
That is one way of calling the regular recon-all stream. You point it to
any input file with the -i flag. Here dcm could be a dicom image, but
can be nifti or another format.
Best, Martin
On 07/01/2013 10:17 AM, Jonathan Holt wrote:
Martin,
I'm going to experiment with data with bet
Martin,
I'm going to experiment with data with better image quality using the
longitudinal stream. In the first step of the workflow I'm confused to what
'dcm' stands for in the following command
recon-all -all -s -i path_to_tpN_dcm
could you elaborate?
jon
On Sat, Jun 29, 2013 at 12:29 A
Hi Jon,
The longitudinal stream in freesurfer is designed to reduce variability in
situations where you have several scans from the same subject. The information,
that it is the same subject, is valuable and by initializing many steps in
recon-all with common data we can increase reliability of
Martin,
I hate to be noobish, but can you elaborate on the longitudinal stream? It may
be that I'm doing entirely too much work.
jon
On Jun 28, 2013, at 11:38 AM, Martin Reuter wrote:
> So you are not processing this with the longitudinal stream? You are just
> running each image independently
my guess is it that it is, as for whatever reason temporal WM seems to
darken first in an array of disorders. I would start with a couple of
control points
cheers
Bruce
On Fri, 28 Jun 2013, Jonathan Holt
wrote:
We have 4 time points for that subject and it's an issue in all 4, to a
lesser
So you are not processing this with the longitudinal stream? You are just
running each image independently through freesurfer and look at the results?
Is there motion in all the 4 time points in this subject or only in the one
image that you send? With motion artifacts, surfaces are not very rel
We have 4 time points for that subject and it's an issue in all 4, to a lesser
degree in one of them, but still an issue. Same area, same hemisphere, every
time. Perhaps it is pathology based, I am not entirely sure.
On Jun 28, 2013, at 11:19 AM, Martin Reuter wrote:
> Hi Jonathan,
>
> is th
Yes, but do these errors show up in the cross sectional data, in the base or in
the longitudinal data (or all?)? Does it look good in the base?
I'd suggest more skull strip where the pail grabs dura.
On Jun 28, 2013, at 11:26 AM, Jonathan Holt wrote:
> Martin,
>
> this is longitudinal proce
oh, if it's that few then yes, but that wasn't what it looked like in the
image you sent
On Fri, 28 Jun 2013, Jonathan Holt wrote:
Martin,
this is longitudinal processing, would you recommend a more aggressive or
less aggressive skull strip?
Bruce,
I've not seen this issue on more than a hand
Martin,
this is longitudinal processing, would you recommend a more aggressive or less
aggressive skull strip?
Bruce,
I've not seen this issue on more than a handful of subjects. They're ALS, but I
dont think that is the cause behind this. I'm unfamiliar with bias fields, but
it's more likely
Hi Jonathan,
is this in cross sectional processing or in longitudinal?
One thing I notice is significant motion artifacts (ringing), which seem to
lead to underestimation of WM, also you could fine tune the skull strip.
Best, Martin
On Jun 28, 2013, at 10:19 AM, Jonathan Holt wrote:
> Hi all
Hi Jonathan
so you lose that entire gyrus? What are the intensity values of the wm in
the brain.mgz in that region? And it's not present in the wm.mgz either?
That means it's a segmentation problem not a topology one (which is what
I would have guessed)
Bruce
On Fri, 28 Jun 2013, Jonathan Ho
14 matches
Mail list logo
