Posted by bob cox on March 19, 2001 at 07:23:05:
In Reply to: Re: 3dDeconvolve for t-test posted by Phoebe on March 18, 2001 at 16:01:43:
> Where can I get the rough idea of the intensity value to be cleared up?
The new program 3dClipLevel can give you an idea of what value
to clip datasets with to exclude non-brain tissue. This was only written
recently, so you'll need an up-to-date version of the AFNI to use it.
If bold+orig is your time series dataset, and sub-brick 20 is the
first one you want to use, then
3dClipLevel -v 'bold+orig[20]'
will print out a value that should work for you. You could also
do something like this (using the C shell csh)
set val = `3dClipLevel 'bold+orig[20]'`
3dcalc -nscale -a 'bold+orig[20]' -expr "step(a-$val)" -output bold_mask
Here, you are capturing the output of 3dClipLevel into the shell variable
val, using the backwards quote operator. Then, this value is being used
in the 3dcalc command, using the $ operator.
bob cox