Hi - good question.....
At the moment, MELODIC produces a set of independent spatial and temporal
maps which are not exactly "inferential" (eg Z-stat or probability spatial
map outputs), but which contain similar kind of information.....
So it is not trivial to make _inference_ at group-level on the basis of
the MELODIC output - however, there is nothing to stop a variety of ways
of combining the results at the group level. The first thing to do is to
put the results in standard space for each subject (see below), then you
can play with different combination methods....we could discuss this
further if useful.
TO STANDARD SPACE:
extract a single volume from the input data as an example fmri:
avwroi input examplefmri 0 1
register to standard space (to find the transform):
flirt -ref ${FSLDIR}/etc/standard/avg152T1_brain -in examplefmri \
-omat <icaoutputdir>/examplefmri2standard.mat
apply this to the 4D analyze file output from ICA (this is what is used to
generate the colour pics on the web page report):
cd <icaoutputdir>
applyxfm4D melodic_IC ${FSLDIR}/etc/standard/avg152T1_brain \
melodic_IC_standard examplefmri2standard.mat -singlematrix
Thanks, Steve.
Stephen M. Smith
Head of Image Analysis, FMRIB
Oxford University Centre for Functional MRI of the Brain
John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
On Thu, 30 Aug 2001, Lukas Scheef wrote:
> Hi all!
>
> I wonder how to perform a group analysis with MELODIC (1st or
> 2nd level)
>
> Thank you very much for your help!
>
>
> Lukas
>
>
> Lukas Scheef, MD, MSc
> MR-Tomography
> Department of Radiology
> University of Bonn
> Sigmund-Freud-Straße 25
> D-53105 Bonn
> Fax: +49-228-287-5598
> Phone:+49-228-287-5960
> E-mail: [log in to unmask]
>
|