Hi
melodic does not remove the mean time course - though you can request this to happen using the --remove_meantc command line option. I suspect that what you see is simply due to the fact that melodic by default performs variance normalisation in order to avoid the tissue-type bias in the PCA step. Have you run melodic without this (i.e. use --vn) to see if that explains things?
hth
Christian
On 18 Jun 2010, at 08:41, Mohamed Bahri wrote:
> Dear FSL experts,
>
> I performed a single-session ICA both with Melodic FSL and BrainVoyager. The visual inspection of the results (Z-map) reveals the presence of one component representing a mean signal (global signal). This global signal component is present only in the BrainVoyager results. So, my question is: does Melodic ICA filter out the global signal? if yes, how does it do that?
>
> Any comment will be very welcome,
>
> Many thanks in advance,
>
>
> Mohamed
|