Dear Christian,
try calling ImCalc from the command line, setting the
datatype to spm_type 'float' like this:
spm_imcalc_ui([],[],'i1.*(i2>3.11)',{[],[],'float',0})
(See also the following link for a complete answer to your problem)
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind04&L=spm&P=R171919&I=-1
Best wishes,
nicola
----- Original Message -----
From: Christian Keysers <[log in to unmask]>
Date: Friday, May 6, 2005 1:52 am
Subject: [SPM] results table for imagecalc T map
> Dear SPM
>
> I often have to do further computations on random effect analysis,
> suchas for instance combining the results of two different tasks
> (e.g. motor
> and visual tasks) to look for overlaps. I do this by doing two
> separaterfx analysis, which create me their own spmT maps. Then
> using image
> calc, I do i1.*(i2>3.11), which gives me the T values of the first RFX
> but only in voxels where the other task was significant as well.
> Not to
> report that in a paper, its nice to have a table. In SPM99 I handled
> that by creating a bogus contrast in the first RFX, look at it,
> and then
> rename my image calc image as spmT003.img/hdr. Then visualizing my
> boguscontrast, I saw my image calc SPMT, and could visualise a
> nice results
> table. In SPM2 that doesn't work anymore, because of the header info's
> etc which are different. Any idea how to get a table from an image
> calcT map?
>
> Christian
>
> --
> Christian Keysers, PhD
> Associate Professor
>
> BCN Neuro-Imaging Center
> University Medical Center Groningen
> Antonius Deusinglaan 2 (room 125)
> 9713 AW Groningen
>
> Phone: +31 50 3638794
> Fax: +31 50 3638875
>
|