[Mrtrix-discussion] CSD data format

[Ariel Rokem]

Hi everyone,

I am trying to reconstruct ODF estimates, based on the results of the CSD
computation. For example, if I follow the instructions given in the online
docs for preprocessing DWI data and end up with a file named CSD10.mif.
After converting this file to .nii format and examining it, I found that it
had a last dimension of size 66, which corresponds to the number of
parameters required to estimate the CSD weights up to harmonic order of 10.
So, I assumed that the 66 values in each voxel in the resulting file were
the estimated coefficients into the set of spherical harmonics used in this
case. Is that correct? If that is correct, as I understand it, all I would
have to in order to reconstruct an odf estimate for a given voxel would
then be to calculate the inner product between the coefficients and the
spherical harmonics basis set (sampled at an array of points on the
sphere). However, I would need to know which coefficient corresponds to
which spherical harmonic function. Is that documented somewhere? Or could
you tell me how these are organized? I have been assuming that they are
organized by harmonic order (0,2,4...) and then by degree or phase factor
(e.g. for order 2: -2,-1,0,1...), but I haven't been able to verify this by
looking at the source-code (I am not really sure where to look).

Thanks,

Ariel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20111209/462ca22c/attachment.html