> I'm attempting to construct a paed T1 template.
> At this stage I intend to use ONLY normal "controls", rather than "controls
> and patients" because many of the patients have multiple areas of cortical
> dysplasia.
>
> There is a total 35 controls for use in template creation AND in any
> subsequent analyses. Is it better to use a subset of these controls (say
> 16- 24) rather than the entire control set of 35?
>
> A subset based template would reduce potential bias in later statistical
> analyses. So the primary question is...what is the minimal number of
> controls need to construct a representative template?
SPM used to use a single-subject template, and many groups still do their
registration to single subject scans. Personally, I don't think that this
is a good idea though.
Ideally, the template would be constructed from as many scans as possible,
although the "law of diminishing returns" applies. There is no hard-and-fast
rule about how many subjects to use.
Using the whole set of scans would be preferable to using a subset. This
can be framed in terms of consistency. See e.g.
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0302&L=spm&P=R55218&I=-1
Best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|