John
> >
> > It is also not necessarily sufficient to encode an orientation. In order
> > to account for imaging gradients etc., you may need a 3x3 symmetric matrix
> > to completely describe the diffusion encoding information.
>
> I'm afraid my lack of knowledge about DTI is letting me down here. I was
> under the impression that people were only after encoding some diffusion
> encoding direction within a volume. The tensors, or whatever else you want
> could then be reconstructed from these individual volumes.
This is usually almost true!! But there are situations where you need more
than an orientation to tell you what diffusion encoding you have. For
example, if you apply equal diffusion-encoding along x,y,z at the same
time, you get diffusion-encoding along orientation [1 1 1], but if you
apply gradients along x, then y, then z in succession, you get _isotropic_
diffusion encoding.
To represent these possibilities, you would need to store a *b-matrix*
associated with each volume. (As mark says, it is 3x3 symmetric).
T
-------------------------------------------------------------------------------
Tim Behrens
Centre for Functional MRI of the Brain
The John Radcliffe Hospital
Headley Way Oxford OX3 9DU
Oxford University
Work 01865 222782
Mobile 07980 884537
-------------------------------------------------------------------------------
|