Has anyone had a chance to take a further look at the output
differences in using the -mprage option yet? (I'm still new at
freesurfer and probably wouldn't know what to look for...)
-Alex
~)---
--
Alexander Li
So, are you saying that the default recon-all stream is really optimized
for SPGR, and anyone running a MPRAGE should be including the -mprage
flag when they run recon-all? If so, this is news to me, and I suspect
many other FS users.
The default recon-all setting is set IsMPRAGE = 0, but I
no, it's more agnostic. Optimized for both spgr and mprage (or neither,
depending on your perspective). You're probably right that mp-rage is more
common these days, but I think spgrs are still fairly widespread on GE.
On Thu, 27 Aug 2009, Michael Harms wrote:
So, are you saying that the
Is there any gain/benefit then by including this -mprage flag? In what
circumstance is that flag to be used?
I'm still confused what the purpose of that flag is.
thanks,
Mike H.
On Thu, 2009-08-27 at 08:54 -0400, Bruce Fischl wrote:
no, it's more agnostic. Optimized for both spgr and mprage
yes, I think it will probably help a bit for mprage scans. Essentially it
changes the default assumptions about noise and contrast characteristics
(it assumes more of both) and will help the intensity normalization get a
bit further out the thin white matter strands. Try it and see what you
Ah I see,
this raises a question then, of whether the optimizations are still
applicable to WU acquired MP-RAGE data, as we are currently collecting
3T Trio data, and I believe Randy's data is 1.5T Vision data, however
the sequence (our CAP MP-RAGE) is quite similar... can anyone up there
if you send us a volume we'll take a look. Basically Randy's mp-rage had
significantly darker gray matter than we were used to in SPGR and MPRAGE
images. If wm was around 110, I think typical GM is around 75, but Randy's
were around 60.
cheers,
Bruce
On Wed, 26 Aug 2009, Alexander Li Cohen
yes, it was optimized for the mp-rages that Randy B used to collect
there, not for example ADNI.
On Tue, 25 Aug 2009, Michael Harms wrote:
I assume that that flag (-washu_mprage) is somewhat antiquated, right?
That is, it was included for an older set of scans (generated at WashU
generated
No, our current CAP MPRAGE at WU on the 3T TimTrio is very similar to
the ADNI mprage, so that old flag shouldn't be used.
-MH
On Wed, 2009-08-26 at 09:24 -0500, Alexander Li Cohen wrote:
Ah I see,
this raises a question then, of whether the optimizations are still
applicable to WU
There's also the flag -mpragewhich can be used, intended for the
mprage developed by Andre van de Kouwe, which I believe is the same as
that now used by Siemens.
Nick
On Wed, 2009-08-26 at 09:37 -0500, Michael Harms wrote:
No, our current CAP MPRAGE at WU on the 3T TimTrio is very similar
Ok, now I'm confused again, because my understanding is that the Andre
van de Kouwe mprage, Siemen's default mprage, and the ADNI mprage really
are not very different in terms of their contrast. Is that not correct?
Assuming they are indeed all similar, are you saying that a modern
ADNI-like
Alex,
We ran into the same problem a while ago for some of our studies. We
did the following:
1. Convert an Avi preprocessed T1 MPRAGE into a FreeSurfer-native
format and provided it the appropriate orientation and image center.
2. Next, we simply ran recon-all on this file (single MPRAGE).
Yes,
I am currently giving freesurfer an atlas (711-2B / TRIO_Y_NDC to be
specific) transformed 111 MP-RAGE file. The concern is that if I use
mri_convert, I want to specifically return to the same 111 volume size
and space and I provided as accurately as possible (hopefully using
the
Alex,
The -washu_mprage flag affects mri_normalize and mri_segment. It
adjusts the parameters of those binaries to match the assumed intensity
profile in mprage scans, eg. to account for darker grey matter. I've
updated the help text in recon-all.
Nick
As a side note, there appears to be a
To: Donna Dierker
Cc: Alan Anticevic; freesurfer@nmr.mgh.harvard.edu
Subject: Re: [Freesurfer] Input/Output to specific atlas space/dimensions?
Yes,
I am currently giving freesurfer an atlas (711-2B / TRIO_Y_NDC to be
specific) transformed 111 MP-RAGE file. The concern is that if I use
I assume that that flag (-washu_mprage) is somewhat antiquated, right?
That is, it was included for an older set of scans (generated at WashU
generated years ago before Siemens had its own MPRAGE sequence) that
happened to have darker gray matter, and doesn't need to be applied for
modern MPRAGE
16 matches
Mail list logo
