If the vectors you put in the sequence are the same as the ones in the dicom
header, and you are angulating the FOV, then everything is how you would
expect. The dcm2nii bvecs have been corrected for the angulation and are
what you should use (and therefore the ones in the DICOM header are
uncorrected and different).
If you were not correcting your bvecs for angulation before, I would have
concerns about those results, even fairly small FOV angulations, when not
corrected for, can give you incorrect tracking results. These differences
may not be apparent unless you are tracking a pathway where the particular
angulation makes a difference and you have a lot of experience with that
pathway previously (for example I have a lot of experience tracking the
arcuate with many different tracking algorithms and acquisitions, and thus
can look at an arcuate result and tell if something is not the way it should
be).
Peace,
Matt.
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
Of Mareike Menz
Sent: Thursday, May 14, 2009 9:17 AM
To: [log in to unmask]
Subject: Re: [FSL] AW: [FSL] bvecs in nifti space of in dicom space?
Hi Matt,
we made the same experience. Nevertheless, the dcm2nii output differs from
the dicom header entries
(and the vectors we put in the sequence) And it does so considerably.
Especially since we often measure angulated.
Before dcm2nii had gave bvecs and bvals, we always used the directions
obtained from our physicist.
That also worked fine.
But since they differ, there can only be one correct one. Am I right?
I do not worry about FA, but tracking should be affected.
Thanks for your help,
Mareike
Matt Glasser wrote:
> I have found the dcm2nii output to work very well and not require any
manual
> adjustments.
>
> Peace,
>
> Matt.
>
> -----Original Message-----
> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf
> Of Andreas Bartsch
> Sent: Thursday, May 14, 2009 3:50 AM
> To: [log in to unmask]
> Subject: [FSL] AW: [FSL] bvecs in nifti space of in dicom space?
>
> Hi,
> probably the latter. I guess the former are in magnet (xyz) not patient
> resp. phase-read-slice coordinates (prs). However, you need to figure out
> yourself to be 100% sure. If you have not angulated your acquisition, the
> values should be the same.
> Cheers-
> Andreas
>
> ________________________________________
> Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag von
> Mareike Menz [[log in to unmask]]
> Gesendet: Donnerstag, 14. Mai 2009 10:14
> An: [log in to unmask]
> Betreff: [FSL] bvecs in nifti space of in dicom space?
>
> Dear FSLers,
>
> we noticed that the bvecs differ in dependence of the readout method.
> An inhouse script reads out the dicom entries as does spm with the helper
> from the diffusion
> toolbox. We already noticed a deviance of these to the gradient table our
> physicist put in but this
> was rather small (maybe due to rounding).
> Now we found out that the dcm2nii output provides yet another gradient
> table. If I understood Chris
> Rorden correctly, this is, because dcm2nii reports the bvecs in
NIFTI-space.
>
> Which one should be used for dtifit, and even more important for bedpostx?
>
> Thanks for your answer,
> Mareike
>
> --
> Mareike Menz
> NeuroImage Nord Lübeck
> University Medical Center Hamburg-Eppendorf, Bldg. S10
> Martinistr. 52
> 20246 Hamburg
> Germany
> +49 40 7410 53683
> [log in to unmask]
> -------------------------
> () ascii ribbon campaign
> /\ - against html mail
>
>
> --
> Pflichtangaben gemäß Gesetz über elektronische Handelsregister und
> Genossenschaftsregister sowie das Unternehmensregister (EHUG):
>
> Universitätsklinikum Hamburg-Eppendorf
> Körperschaft des öffentlichen Rechts
> Gerichtsstand: Hamburg
>
> Vorstandsmitglieder:
> Prof. Dr. Jörg F. Debatin (Vorsitzender)
> Dr. Alexander Kirstein
> Ricarda Klein
> Prof. Dr. Dr. Uwe Koch-Gromus
>
>
--
Mareike Menz
NeuroImage Nord Lübeck
University Medical Center Hamburg-Eppendorf, Bldg. S10
Martinistr. 52
20246 Hamburg
Germany
+49 40 7410 53683
[log in to unmask]
-------------------------
() ascii ribbon campaign
/\ - against html mail
--
Pflichtangaben gemäß Gesetz über elektronische Handelsregister und
Genossenschaftsregister sowie das Unternehmensregister (EHUG):
Universitätsklinikum Hamburg-Eppendorf
Körperschaft des öffentlichen Rechts
Gerichtsstand: Hamburg
Vorstandsmitglieder:
Prof. Dr. Jörg F. Debatin (Vorsitzender)
Dr. Alexander Kirstein
Ricarda Klein
Prof. Dr. Dr. Uwe Koch-Gromus
|