On Mon, 29 Aug 2005, Ray Fix wrote:
> Step 1: Run the optimized VBM preprocessing script.
>
> Step 2: Think of a statistic you are interested in looking... e.g. score on
> some test.
This is a crucial point. When modelling/grouping your data with a linear
model, you need to model all necessary explanatory variables (scores,
group membership...) together in one design. You can then test influences
of the columns of your design using t- or F-contrasts, once the model is
estimated.
> Step 3: Make a new directory identifying that experiment.
Thats allright. This should become the directory where your SPM.mat,
betaXXXX.img's and any other statistics related files will end up.
> Copy all 50 smoothed grey matter images in.
Not necessary - keep them in a safe place out of any analysis directory.
Before specifying the model, just cd into your new directory and start
model specification from there.
> Step 5: Create a basic model (multiple regression with constant). Carefully
> select images to match the order of X. Make the first column X(:,1) control,
> X(:,2) non-control X(:,3) interesting statistic, up to 4 columns.
See above - the basic models might be inappropriate for larger and more
complicated designs, there is more choice if you turn to the PET model section.
> If you have to do something simple, like add one new subject to the study, it
> literally takes weeks to redo because there is so much manual interaction.
> (Normally you should be happy when you get new data, right?) It would be
> ideal if we could just add (or remove) subjects or tweak one aspect of the
> process and turn the crank.
Well, with another subject being entered, you will have to re-specify the
model. And for basic models/PET models even in SPM5 there is no easy way
to auto-configure this.
Yours,
--
Volkmar Glauche
-
Department of Neurology [log in to unmask]
Universitaetsklinikum Freiburg Phone 49(0)761-270-5331
Breisacher Str. 64 Fax 49(0)761-270-5416
79106 Freiburg
|