Hi,
FIRST does not include a TIV measure in the vertex analysis.
One option to normalise the results is to do the vertex analysis in MNI space (--useReconMNI), as then each subject has been registered to the standard space and this consequently normalises for brain size. So it is not really necessary to do more normalisation in this case unless you believe that (i) TIV and brain size will not be highly correlated in your sample (which would be unusual), and (ii) that it is preferable to normalise by TIV rather than brain size, which I don't think there is a clear consensus on, and is likely to be different in different pathological conditions.
Another option is to use a covariate of either brain size or TIV in your statistical analysis. You can use measures derived in any way for this (FSL, SPM, etc.) as they are just entered into the appropriate EV as simple numbers.
Some people may consider that the first option is the "automatic" normalisation, but as explained, it is not the same as automatically including a TIV covariate.
All the best,
Mark
On 6 Nov 2012, at 12:43, Nils Opel <[log in to unmask]> wrote:
> Hey all,
>
> since i could read in some articles that First Vertex Analysis was performed including total intracranial volume (TIV) as a covariate...
>
> Does FIRST Vertex Analysis "automitically" include the TIV of each Subject?
>
> If not so, does it make sense, is it recommended or is it even necessary to insert total intracranial voume as a covariate when running FIRST Vertex Analysis?
>
> If this is the case, is it legitimate to use SPM generated data for TIV or would you recommend other tools for TIV calculations?
>
> Thanks a lot in advance!
>
> Best regards,
>
> Nils
>
|