Hi Markus
> does anybody know whether realignement of epi data in nifti format in spm5
> can take much more time than the old analyze version. only registering the
> images (around 300) takes more than 120 minutes, compared to 15 minutes
> with the analyze format on the same computer. i thought, nifti would be much faster.
> could it be due to changes of the defaults (6th instead of 4th b-spline
> interpolation, quality 1 instead of 0.9, 2 mm separation instead of 4)?
The increase in computation time is definitively due to the changes in the realignment settings.
By changing the separation from 4mm to 2mm, you use 8x as many image points.
Increasing the quality from .9 to one means, you don't use all voxels, not discarging those that contribute little to the realignment (based on the determinant of the inverse covariance matrix)
And on top of that, you also switched to a higher order interpolation, which means that each individual interpolation will take a little longer as well.
So going from 15 to 120 minutes is well within range of what you would expect and there is no need to worry
Cheers
Simon
>
> by the way: what is the differences of 4d nifti(all volumes in one
> file) or
> 3d nifti (n single volumes) for working with spm5? if 4d is
> better, how can
> i convert 3d files into one single 4d file?
>
> thanks a lot
> markus
>
-----------------------------------------
Simon Eickhoff
Institut for Medicine (IME); Research Center Jülich
Leo-Brandt-Str. 5; 52425 Juelich, Germany
Phone + 49 2461-61-5219 / Fax + 49 2461-61-2820
e-mail: [log in to unmask] / [log in to unmask]
|