> As I understand that MRI abnormalities are special cases where we can switch
> off the step of warping.
Optionally, yes. You could go with an affine-only-normalization, which
may be more robust but will also result in a lesser agreement between
the normalized images. I would not recommend that.
> Please, could you give some details about that, ie. for example some works
> that neglect this step. Because I must justify this.
The recent paper by Jenny Crinion (NeuroImage 2007, 37: 866-675) may be
a starting point as she investigated the "traditional" cost-function
masking approach with the segmentation approach as implemented in SPM5.
Overall, the latter was actually performing better.
> In fact, I don't understand the warping step in SPM 5. Sorry.
In that case, the SPM manual and the papers listed in the bibliography
(www.fil.ion.ucl.ac.uk/spm/doc/biblio/Keyword/NORMALISATION.html) may be
good starting points.
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473