This is exactly what fsl_motion_outliers
helps you do. film_gls begins to struggle with large design matrices (if
you have high resolution and low TR data especially), but this depends on how
much RAM you have.
Peace,
Matt.
From:
Sent: Monday, May 07, 2012 11:40
PM
To: [log in to unmask]
Subject: Re: [FSL] Censoring
Timepoints
Hi
On 8 May 2012, at 03:48,
Hello All-
I'm wondering if anybody has a good method, staying within FSL's armament of
tools, to perform analyses with select TRs "censored" in the way that
AFNI's 3dDeconvolve and 3dREMLfit can do. I've had some thoughts about adding
an confound EV of 0s and 1s describing TRs that I do/do not want associated
with my regressors of interest,
Yes - that is a reasonable approach - as long as your design matrix
does not get impractically large! You need one EV for every timepoint to
be "removed".
Cheers.
but that isn't exactly a
replication of AFNI's approach.
Any ideas?
Thanks,
Jim
------
Jim Porter, M.A.
Ph.D. Candidate
Clinical Science and Psychopathology Research
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,
FMRIB, JR Hospital, Headington,
+44 (0) 1865 222726 (fax 222717)
[log in to unmask]
http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------