Hi Olivier,
The "correct header information" can be harder than it sounds...
I think the notes, m-files, and maybe .hdr files at the following
location should help you out (I haven't made the .img or .rawb files
downloadable, as I don't expect I have permission to redistribute
BrainWeb, but if you download the rawb from the main BW site you
should be able to reproduce my results)
http://www.cs.ucl.ac.uk/staff/gridgway/bw_nifti/
Note that the original BrainWeb and the 20 new models are different
sizes, also, for the BW20 models, the segmentations/discrete-model are
different sizes to the grey-level images -- I've put up m-files for
handling all of these, but take care.
Note also, I've just done this for SPM5 NIfTI (they are two-volume
hdr/img NIfTIs rather than single-volume .nii, but don't be fooled
into thinking they are Analyze). With Analyze, all sorts of weird
storage-order world-left-right flipping nonsense occurs, which I
haven't attempted to deal with. You might find the easiest route is to
get NIfTI pairs which you are confident are correct, and then convert
these to Analyze in a way which appears to work with your SPM2 set-up
(including the defaults.analyze.flip business).
All the best,
Ged.
Olivier Salvado wrote:
>
> I am using the BrainWeb datasets from McGill to test different versions
> and parameters of the segmentation step.
> (_http://www.bic.mni.mcgill.ca/brainweb/_)
> I have downloaded the datasets in raw format, and written a function to
> read them in matlab. Now I need to go through all the volumes and
> generate the nifti and analyze files to import them into SPM (SPM2 and
> SPM5) with the correct header information.
>
> Has anyone already done that and could share the converted volumes with me?
>
> Olivier
>
> __________________________________________________
> Olivier Salvado, PhD
>
> Case Western Reserve University | Case Center for Imaging Research
> University Hospitals of Cleveland | Department of Radiology | Wearn B49
> 11100 Euclid Av. | Cleveland, OH 44106
> Ph. (216) 983 3426 | Fax: (216) 844 4987
>
>
>
>
|