Hi - I think this should be fine. It sounds like you did independent
ICA analyses for each subject. Because the DMN timecourse (that you
presumably got from melodic_mix) has stddev=1, you can indeed compare
these copes across subjects, and your higher-level setup looks fine.
Cheers, Steve.
On 11 Mar 2008, at 23:14, Vina Goghari wrote:
> Dear FSL experts,
>
> I would be very grateful for some help on a resting state fmri
> analysis we are trying to do. We want to compare the resting state
> data from two groups (29 clinical and 29 control) while also looking
> for correlation of BOLD signal in the default mode network during
> resting state and behavioral/cognitive measurement.
>
> So far we have done this:
> 1. Preprocessed data (motion correction, smoothed, filtered, etc)
> 2. Denoised each individual data (motion, heart, respiration)
> 3. Put all the data in Melodic and got 35 ICA components
> 4. Identified the component that corresponds to the default mode
> network
> 5. Used timeseries of default mode network ICA component as a costum
> waveform to do voxelwise correlation in each individual's dataset
> (created cope images in first level analysis).
>
> In the higher level analysis we would like to:
> 1. Compare cope images between groups (two sample unpaired t-test?)
> 2. Investigate additional effect of behavioral measures gotten from
> cognitive batteries (WAIS, WMS, etc) for each group and compare that
> between groups (as in single group average w/additional covariate???)
>
> I have the Full model setup like this, but I am not sure if it is
> right (5 subjects per group to make this a little shorter):
>
> EVS:
> group EV1 EV2 EV3 EV4 (group
> A) (group B) (score-group A) (score-group B)
> 1 1 0 -1 0 1
> 1 0 -3 0
> 1 1 0 2 0
> 1 1 0 -2 0 1
> 1 0 -1 0
> 2 0 1 0 1
> 2 0 1 0 2
> 2 0 1 0 3 2
> 0 1 0 2
> 2 0 1 0 -1
> Contrasts and F-tests:
> Title EV1 EV2 EV3 EV4
> C1 group A mean 1 0 0 0
> C2 group B mean 0 1 0 0
> C3 group A>B 1 -1 0 0
> C4 group B>A -1 1 0 0
> C5 group A score 0 0 1 0
> C6 group B score 0 0 0 1
> C7 group A>B score 0 0 1 -1
> C8 group B>A score 0 0 -1 1
>
> Can you please let me know if this is correct? Do I need to
> orthogonalize? I am new to FSL and would appreciate any kind of
> help. Thanks!
>
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|