Hi - that looks sensible,
Cheers.
On 29 Aug 2009, at 01:22, Julius Fridriksson wrote:
> Thanks for your feedback Liam. Here is what I was also considering:
> The
> patients show a wide range of benefit from our treatment. To
> understand what
> areas predict outcome, I was planning to do a t-test for each
> individual
> first ((Session 1 + Session 2) - (Session 3 + Session 4)) using a
> fixed
> effect comparison. Then, in the next level analysis, to input the
> resulting
> COPEs to predict group outcome (e.g. greater signal change = greater
> recovery) using mixed effects. That is, are there specific areas
> that show
> greater signal change in those patients that show greater recovery?
> Again,
> hope this makes sense. -
> Julius
>
> The way I understand your analysis is as follows:
>
> 21 subjects in the before treatment condition
> (same) 21 subjects in the after treatment condition
> You want to compare the two conditions - this you do with a paired t-
> test
> within a Mixed effects analysis
>
> Say you have two runs for each condition? You would average across
> these
> runs using a Fixed effects analysis before using the COPE files of
> the Fixed
> effects analysis to do the Mixed effects paired t-test.
>
> Hope this makes sense.
>
> Liam.
>
>
>
> *******************************************
> Liam Nestor, Ph.D
> Office C8-523
> Laboratory for Molecular Neuroimaging
> Semel Institute for Neuroscience and Human Behavior
> 760 Westwood Plaza
> Los Angeles 90024
> Tel: 310-206-0655
> E-mail: [log in to unmask]
> *******************************************
> ________________________________________
> From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of
> Julius Fridriksson [[log in to unmask]]
> Sent: Friday, August 28, 2009 12:59 PM
> To: [log in to unmask]
> Subject: [FSL] higher level FEAT
>
> Hi - I have a group of 21 patients who underwent two fMRI sessions
> before
> and after treatment. I would like to compare activation before and
> after
> treatment on patient-by-patient basis and then do a higher level
> regression
> analysis to see whether change in group activation predicts
> behavioural
> improvement. For the within subject analysis I am using a higher level
> t-test comparing the two before and two after fMRI sessions. My
> questions
> is: Should I use a fixed effect analysis for this step?
>
> Also, for the higher level (third level) analysis that includes the
> whole
> group, should I use the COPEs as inputs? Hope this makes sense.
> Thanks -
> Julius
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|