Print

Print
David,
Hello. I dont think the zero matrix should work. The design matrix is needed by the film program because it is calculating the residuals, so it needs to regress out something and calculate the autocorrelation.
So a model fit to a zero matrix should not work. The ramp matrix works because it will result in a fit and even if this ramp is removed from the data, it should not matter, unless your scientific question is related to linear drift.

Jason.

On Thu, Feb 18, 2010 at 3:27 PM, David Ruhl <[log in to unmask]> wrote:
Hi Jason,
Is this preferable to an all zeros matrix (which I was using for this application) for some reason?
Thanks
-David


On Thu, Feb 18, 2010 at 12:35 PM, Jason Steffener <[log in to unmask]> wrote:
Dear Leo,
I have done what you are interested in. I created a RAMP design matrix with FEAT and used that as the design.

To create the RAMP design I ran fsl>feat and entered the number of time points I had N, then for the design I selected "custom 1 entry per volume." This was a text file containing the values 1 to N.

My final command was:
fiml_gls -rn PreWhiteStats -output_pwdata DATAFILE Ramp.mat

Jason.



On Wed, Feb 17, 2010 at 7:52 PM, Yiou Li <[log in to unmask]> wrote:
Hi Stephen,

Thanks for your answer. When I run film_gls on my data, it says

"An exception has been thrown
Paradigm filename needs to be specified."

Can I turn off this requirement and just run temporal whitening on the raw data?

If this requirement can not be turned off, since my data is resting
state, what design matrix should I use?

Best,
Leo

On Mon, Feb 15, 2010 at 1:17 AM, Stephen Smith <[log in to unmask]> wrote:
> Hi - flm_gls can output the whitened data with the -output_pwdata flag - you
> would want to apply this to the filtered_func_data created by FEAT.
> Cheers.
>
>
>
> On 11 Feb 2010, at 19:40, Yiou Li wrote:
>
> Hi all,
>
> I am wondering if there is a handy command in fsl to perform temporal
> whitening on fMRI data (to remove the temporal autocorrelation)? It
> seems FEAT should have such feature but I can't find it explicitly
> either from the FEAT GUI menu or FEAT command help.
>
> Comment is really appreciated!
>
> Leo
>
>
>
> ---------------------------------------------------------------------------
> 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
> ---------------------------------------------------------------------------
>
>
>
>



--
Jason Steffener, Ph.D.
Department of Neurology
Columbia University
http://www.cumc.columbia.edu/dept/sergievsky/cnd/steffener.html




--
Jason Steffener, Ph.D.
Department of Neurology
Columbia University
http://www.cumc.columbia.edu/dept/sergievsky/cnd/steffener.html