Hi Jokel
> Dear SPM-experts!
> I extracted a timecourse/signal (first eigenvariate) from a couple of
> regions of interest from a blocked design. However I would like to
> remove the task-related activation from this signal and there seemed
> to be to approaches:
> 1. Simply from the signal of each block substract the mean signal of
> this block.
> 2. Deconvolve the signal by modelling task-related signal-changes with
> the help of the hemodynamic response function and regressing out
> task-related activation.
> The first approach is of course quite easily scripted in matlab,
> however in the second case I would be very thankful for some hints how
> to implement this approach in matlab and with the help of spm's functions.
I'm not sure this is the right way but the basic deconvolution in matlab
is to use the deconv, in your case with the hrf generated using spm_hrf
function.
Within spm the deconvolution is much more sophisticated (used for DCM)
as it is done via inversion of the forward model that generate the BOLD
(see DCM papers), but the deconv function might do the trick, depending
on what you wanna do ..
good luck
cyril
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|