Hi Russ,
> I would like to make an argument in favor of using the adult template
> for normalization of kids' data, and against using age-specific
> templates. If you use different templates to normalize different groups
> of kids, then there will be systematic differences between the age
> groups and your comparisons between age groups are necessarily
> confounded with anatomical differenecs.
I fully agree: if one wishes to make inferences about different age
groups in one study, one should use the same template for all of them,
for exactly the reason you specified. However...
> If you use the adult template
> then there will be increased noise in the normalization, but since the
> target is the same for all subjects the age comparison is not
> necessarily confounded.
This is not only true if the adult template is used but also if ONE
pediatric template is used for each study (I hope I never advocated
using different templates within one study). In addition, a pediatric
template will be closer to the group under investigation in shape and
tissue composition (most important for an optimized processing stream
and if young children are included), which is why I think that using a
pediatric template still beats using an adult one.
> Also, note that Steve Petersen's group has published a couple of papers
> in NeuroImage (Kang et al., 2003; Burgund et al., 2002) which have shown
> that normalization of kids' data to an adult template actually does
> quite a good job.
I have read these and I agree that the results look very decent.
However, when I compare my results from both approaches, I very much
favor the pediatric reference data.
Best from Tuebingen,
Marko
--
=====================================================================
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
=====================================================================
|