I actually had the same problem, it can be quite tricky to solve when the
scans differ between runs. I created a set of Linux shell scripts that uses
FSL's avwutils to count the number of scans in each file. Then it creates a
batch file from a template (a manually created batch file with all image
references deleted) and runs a loop that adds a new image line for each
image count. Then it starts Matlab and runs the batch.
A much simpler solution would have been if you could instruct SPM to map all
scans in a 4D file, but unfortunately such a functionality is not
implemented. Thinking of it, it should be quite easy to write a simple Java
program that handles this. Perhaps I will do that instead if I come across
this problem again.
I'd be happy to send you my scripts if you like to have a look at it.
On Wed, 11 Apr 2007 21:45:11 +0100, Claus Lamm <[log in to unmask]> wrote:
>Using SPM5, I set up a default preprocessing batch-script that specifies all
>the preprocessing steps.
>now I want to run those steps for all my subjects, i.e. basically run a loop
>I saw several entries on the helpline but weren't really able to implement
>this. one problem might be that my study has several runs, with different
>number of scans per run.
>so my question is: how could I implement such a loop? or, alternatively: how
>can I change the data/filenames in my batch.mat?
>THANKS in advance