Hi,
On 30 Jul 2007, at 04:59, Vinod Venkatraman wrote:
>> No, sorry. However there is a danger is re-using parts of previous
>> analyses, which is that you end up with a possibly messy set of
>> directories and files, and end up getting confused. I generally
>> recommend always running FEATs from scratch, even though that takes
>> longer to run, as things end up neater, which in the long run is very
>> important.
>
> While I agree that it is necessary to have the directories
> consistent across
> all sessions, I would assume one solution would be to allow to
> specify the
> mat files from the first time you have run the analysis as inputs
> to flirt
> subsequently. This would save a lot of times and allow people to run
> multiple models on the same dataset, and also keep all directories
> consistent. Is this right?
In theory that could be done but adding that kind of functionality
would increasingly make the FEAT GUI and scripts over-complicated
both from the user's point of view and the programmer's - and FLIRT
is actually pretty fast anyway so it's better to keep things simple.
> Also, is there any other danger (apart from directories
> consistency) in
> turning registration off for subsequent analysis and then copying
> the 'reg'
> directory from the previous analysis? From looking at my log files, it
> appears that the only other thing FSL does after registration at
> first level
> is clustering of zstat images, which can also be run using a shell
> script
> subsequently. I am really pressed for CPU processing time and want
> to try
> 3-4 different models on 20 subjects and hence would like to save on
> any
> processing time possible.
Sure - if you want to script this yourself as a post-first-level-FEAT
step that's fine. Unless you need your cluster-tables to report in
standard space you don't need to worry about the re-running the
cluster stuff. In fact in the new release the different FEAT stages
are fully independent script calls so it will be easy (for example)
to re-run the clustering stuff after fixing reg in such a custom way.
> I also have another related question. If I have run my first level
> analysis
> using a set of contrasts and want to add a few more contrasts, is
> there a
> simpler way to do this without having to go through all the
> analysis again?
> As a simple example, if I have Evs A, B and C and have already
> created cope
> images for A > 0, B > 0, C > 0 and A > B and now want to run the
> analysis
> for A > C, is there any way I can do this in matlab by just using
> the pe,
> cope and varcope images for A and B from my previous analysis? Or
> would I
> have to run the entire first level analysis all over again?
Sure - you can even do this from the GUI - just change the top-right
pulldown menu.
Cheers, Steve.
------------------------------------------------------------------------
---
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
------------------------------------------------------------------------
---
|