Hi,
I would like to adapt this standard data scaling code to a dataset that has
multiple rows with one row for each voxel, and multiple columns representing the time axis.

3dcalc -a dataset.1D -b mean.1D	    \
-expr 'min(200, a/b*100)*step(a)*step(b)'  \
-prefix dataset.scaled

Unfortunately, I could not find a way to trick 3dcalc to treat these files as voxel time series and the code above gives an error.
I assume "1deval" has to be used, but I am not familiar enough with the syntax. Could you help me writing the code that does the following? :
-loop over each row of dataset.1D
-calculate the expression above with "1deval" (each row has to be transposed first, since 1deval only accepts columns)
"-a dataset.1D -b mean.1D"
-if necessary, remove the headers the .1D files before calculating.
-glue all outputs together again with 1dcat.
-transpose the result again, since 3dDeconvolve would like one row per voxel.

Or maybe you have an idea for a simpler solution. Thanks a lot!



Edited 1 time(s). Last edit at 02/11/2015 09:18AM by rick reynolds.