Dear Matt, Nathan,
think you guys 'simply' have to make a GLM model for a 'simple'
substraction analysis (do you mean Donders type substraction, aplied on
time series data?), and then express your substraction in a post-hoc
contrast, and decide upon the thresholding approach to use (FWE,
FDR,etc), imcalc is not the tool to use here. Activation data is usually
acquired in the form of 100s of images, and you try to model these, and
you substract the fitted amlitudes of your model, not simply images.
It seems that what you want is actually not as simple as you think, you
might want to get into fMRI statistics before you start doing things.
When this is indeed what you want, I'd suggest to read the introduction
of the human brain function 2 book, and especially the chapters on the
general linear model. See:
http://www.fil.ion.ucl.ac.uk/spm/doc/books/hbf2/
Good luck,
Bas
--------------------------------------------
Dr. S.F.W. Neggers
dept. of Psychonomics,Helmholtz Institute
Utrecht University
Heidelberglaan 2
3584 CS, Utrecht, room 17.09
the Netherlands
Tel: (+31) 30 253 4582 Fax: (+31) 30 2534511
E-mail: [log in to unmask]
Web: http://www.fss.uu.nl/psn/pionier
--------------------------------------------
Op di, 29-03-2005 te 10:17 -0800, schreef Matthew Tinsley:
>
>
> ______________________________________________________________________
>
> I'd actually love to get the same set of instructions if anyone was
> planning on replying to Nathan directly.
>
> Thanks, Matt
>
> Begin forwarded message:
>
> From: Nathan Klimenko <[log in to unmask]>
> Date: March 29, 2005 10:11:33 AM PST
> To: [log in to unmask]
> Subject: [SPM] subtraction analysis
> Reply-To: Nathan Klimenko <[log in to unmask]>
>
> Dear List,
>
> I would like to do a simple subtraction analysis using the
> spatially
> normalized images from two groups.
>
> Using Imcalc, I got the average-images of the two groups and
> substracted
> them to get the activation/deactivation images. Would you mind
> showing me
> how to do the following with Imcalc?
>
> 1)select a threshold for masking and exclude all voxels below
> that
> threshold.
>
> 2)Calculate the variance across all voxels and divide the
> masked
> activation/deactivation image by the variance.
>
> 3)Exclude areas with unusually high variance.
>
> Thank you,
>
> Nathan Klimenko
>
>
>
> "The more the universe seems comprehensible, the more it also seems
> pointless." Steven Weinberg (1977) The First Three Minutes.
--
|