Hi Steve,
Thanks for your advice, a straightforward approach and I get the point.
However, wondering if I could set up a joint design matrix for all
subgroups (in GLM setup as different explanatory variables) and then
create different contrast testing for different subgroups (and
possibly other contrasts), and put these contrasts into the randomise
process? Would there be any difference in the statistics for these
tests compare to your suggestion?
Cheers,
Finn
On Sat, May 30, 2009 at 10:19 AM, Steve Smith<[log in to unmask]> wrote:
> Hi,
>
> You can just use fslroi to extract the relevant timepoints (i.e. subjects)
> from all_FA_skeletonised - and then create appropriate design.mat and
> design.con files for testing within subgroups using randomise. This is
> probably the simplest approach.
>
> Cheers.
>
>
> On 27 May 2009, at 15:03, Finn Lennartsson wrote:
>
>> Hi,
>>
>> Just a quick question.
>>
>> We have a control and a patient group. We have done TBSS stats on the full
>> patient
>> group vs control group. Now we want to do sub group TBSS analysis, with
>> defined
>> patient subgroups vs the control group.
>>
>> At which step in the tbss-pipeline do we have to start from?
>>
>> Is there someway the automatically created folder structure from tbss_
>> processes can
>> be changed (e.g. creating /stats_X instead of /stats)?
>>
>> Cheers,
>> Finn
>>
>
>
> ---------------------------------------------------------------------------
> 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
> ---------------------------------------------------------------------------
>
|