Hi Andreas,
This is not terribly clear to me. I'm not sure what they mean by using the b-matrix to calculate the directions, unless they mean reading the private data tags as we are already doing. We don't have this WIP, so I can't test it myself. If you run MRIConvert as it is, are the bvecs what you expect?
Jolinda
--
Jolinda Smith
MR Physicist
Lewis Center for Neuroimaging
University of Oregon
Eugene, OR 97403
[log in to unmask]
On Tue, 3 Jul 2007 09:55:26 +0200, <[log in to unmask]> wrote:
> Hi,
> I traced that to the roots at Siemens, and the message is that if you specified prs in your DiffusionVectors.txt you should use the gradient directions according to that file. At the moment, there is no easy way to recover them from the DICOMs - except if, and that would be up to the converters and in this case you, Jolinda;), if you use the b-matrix to calculate the directions. However, this also contains the spoiler gradient and the back rotation may be flawed in some instances.
> Bottom line - use xyz or, for prs, the entries in the DiffusionVectors.txt.
> Hope that helps-
> Andreas
>
> ________________________________
>
> Von: Bartsch, Andreas
> Gesendet: Do 28.06.2007 07:19
> An: [log in to unmask]
> Cc: 'Jolinda Smith'
> Betreff: WG: [FSL] DTI direction table
>
>
>
>
> Hi Jolinda,
>
> This is in response to your FSL mail, probably better to post it here.
>
> >These are relative to the scanner
>
> Not always: If you define:
>
> CoordinateSystem = prs
>
> in your DiffusionVectors.txt file, they will be relative to the image. Does MRIConvert account for that?
> Thanks,
> Andreas
>
>
> -----Ursprüngliche Nachricht-----
> Von: FSL - FMRIB's Software Library [mailto:[log in to unmask]] Im Auftrag von Jolinda Smith
> Gesendet: Donnerstag, 28. Juni 2007 00:51
> An: [log in to unmask]
> Betreff: Re: [FSL] DTI direction table
>
> The directions are stored in the CSAImageHeader private tag, under the heading DiffusionGradientDirection. These are relative to the scanner, not the image, so they will need to be corrected for any rotations in your slice prescription. Our program MRIConvert will do this automatically, if you want to try it. You can download windows or linux versions at lcni.uoregon.edu/~jolinda/MRIConvert.
>
> --
> Jolinda Smith
> MR Physicist
> Lewis Center for Neuroimaging
> University of Oregon
> Eugene, OR 97403
> [log in to unmask]
>
> On Wed, 27 Jun 2007 11:18:39 +1000, Fang Ke <[log in to unmask]> wrote:
> > Hi, Dear all:
> > We used the SIEMENS TIM 3T MRI system to get the DTI images. Now
> > we have a prolbem. We want to get the gradient direction table from
> > the DICOM files. How can we get it and what data type it is?
> > Thanks a lot.
> > Best regards
> > Ke
> >
> >
> >
> > --
> > Ke Fang
> > Research Officer
> > Neuroimaging and Neuroinformatics
> > Howard Florey Institute
> > University of Melbourne Vic 3010
> > Ph: 61-3-8344 6510
> > Fax: 61-3-9347 0446
> > E-mail: [log in to unmask]
> >
> >
> >
> > --
> > Ke Fang
> > Research Officer
> > Neuroimaging and Neuroinformatics
> > Howard Florey Institute
> > University of Melbourne Vic 3010
> > Ph: 61-3-8344 6510
> > Fax: 61-3-9347 0446
> > E-mail: [log in to unmask]
> >
>
>
>
|