Steve,
Great! Thanks for the heads up.
-Meredith
Quoting Steve Smith <[log in to unmask]>:
> Hi Meredith,
>
> Sorry not to answer your question directly - but I would just recommend
> waiting for the new release of FSL, hopefully quite soon - as this has
> proper multi-subject ICA - so this is probably the best route forward
> for you - hope that's ok?
>
> Cheers, Steve.
>
>
>
> On 31 Jul 2007, at 01:29, Meredith Braskie wrote:
>
>> Hi All,
>> I've been reading through past threads about using melodic at a
>> group level and just want to
>> make sure I understand what needs to be done, as I am new to using
>> melodic like this. We have a
>> number of subjects, each of whom has 6 functional scans of the same
>> task. We are interested in
>> determining which areas on average co-activate with a single region
>> of interest, and also want to
>> investigate whether the ICAs associated with that ROI will vary
>> with another continuous variable
>> (such as age) across subjects. Based on past threads, I think we
>> need to do the following:
>> 1) Run melodic on each individual scan.
>> 2) Select ICAs that include our region of interest (except for
>> those that are obvious noise).
>> 3) Do avwsplit and avwmerge to merge those ICAs of interest into a
>> single 4D file for each scan.
>> 4) Move all the functional data into common space using flirt to
>> create the transformation
>> matrix and applyxfm4D to apply it.
>> 5) Run flame -cope=4Ddata -vc=4Ddata -mask=mask -ld=stats -dm=design.mat -
>> cs=design.grp -tc=design.con -jols
>>
>> Is this correct so far?
>>
>> My questions are:
>> 1) Since we obviously did not do a first level FEAT with the task
>> paradigm outlined in the stats,
>> what exactly will be the output of the flame? Will it find the
>> average of the ICAs? Does it matter if
>> there are different numbers of ICAs for each scan?
>> 2) Once we have run that second level flame (one for each subject)
>> as -jols, will the correct files
>> (i.e. copes etc.) be created so that we will be able to perform the
>> flame across subjects in the gui?
>> If not, how would we proceed to the next step? Will we be able to
>> see how certain ICAs are or are
>> not present on average as a continuous variable changes across subjects?
>> 3) I saw that some threads talk about using the Zstats images.
>> Should we be using those
>> instead of the 4D concatenated files to do what I want to do? If so, how?
>> 4) In the command applyxfm4D that was suggested for bringing the 4D
>> file into common space,
>> what is the -singlematrix for?
>>
>> Sorry for the long email. I just want to make sure we get it right.
>> Thank you for your help with this!
>> Best,
>> Meredith
>
>
> ---------------------------------------------------------------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
> +44 (0) 1865 222726 (fax 222717)
> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
> ---------------------------------------------------------------------------
|