Hi Katrien,
thanks for your mail. I copy to the list as the question pops up now and
then.
> My analyses are going very well, and I have a lot of data (as most
> neuroimagers, i guess). But I do have a question and I wondered if you
> could help me?
Better too much data than not enough ;)
> I calculated an average brain from five T1's of my children to present
> the data on (via MCalc in spm).
Why not all of them? Even if it does get more blurred, that is just how
your data is.
> In order to correctly label the
> significant activations with their name and BA region I use the AAL
> template and BA template in MRICro.
Don't do that, it's not going to work. Two reasons: one, BA maps are
based on adult brains studied in the 19th century, and to beleive in
them for adults is already a leap of faith in my eyes, giving the
enormous variability in brain microstructure (see, eg, "Brodmann's areas
17 and 18 brought into stereotaxic space-where and how variable?" Amunts
K, Malikovic A, Mohlberg H, Schormann T, Zilles K, NeuroImage 2000).
Second, they have been mapped to the adult single subject brain, which
is generally in the same overall space as the pediatric template, but
correspondence between regions must be expected to be poor as kids
brains differ from adult brains in shape and tissue composition (much
less so in size, btw).
I would think that using any customized template makes the application
of the Talairach daemon or the aal maps less correct, but for my kids
data, I never used it and do not recommend doing so.
> but I thought that using the paed. template in normalisation put my data
> within the talairach space so that a coor dinate x y z should be the
> same.
Remember SPM uses MNI-space, NOT Talairach space (for a comparison, see
http://www.mrc-cbu.cam.ac.uk/Imaging/Common/mnispace.shtml). So you
would have to do a conversion anyway.
> In other words, if i find an activation 34/-36/2 (hippocampus) in
> the literature on an adult population and i find an activation in
> 34/-36/2 in my data it should still be the hippocampus? is this
> assumption wrong? because it seems to me that the template is not
> exactly the same to a standardized spm adult template.
Not sure, Yes, and yes. To see if your activation is in the hippocampus,
just overlay your activation map on your averaged anatomical and see for
yourself. No better proof than putting your results and an anatomy
texbook side by side can be obtained. While I see the virtue of putting
numbers to activation, I also think that we have to admit when these
numbers (quite literally) are out of place.
> Or maybe I am too
> hung up on the MRICro templates for naming the regions/BA's and they are
> not completely accurate? Or it was not a good idea of me to use an
> average anatomical and i should use the template as in the package?
Not sure, probably not, and no.
> In any case, I would like to find a means to label the regions correctly
> (since i am not a neuroradiologists, i don't trust my 'hunch'). Could
> you maybe clarify this a little
I beleive that a good hunch may be better than a bad, but scientific
looking label.
> Thank you very much for helping me out - once again :-)!
Oh well, that's what I am here for (among other things, I must admit ;)
Best,
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
=====================================================================
|
