I think the whole problems stems from the radiological vs. neurological coordinates.
My goal is to display AFNI preprocessed data on the MNI 152 template.

To make it short: would the correct solution to the problem be to simply extract voxel-based data using 3dmaskdump via adding the option

-nbox 3 3 3

to 3dmaskdump?

As I understand, -nbox 3 3 3 would extract the voxel-based data in the neurological space matching the MNI152 template, and the 3 3 3 specification would match AFNI’s 3x3x3 mm functional and preprocessed voxel size that I have.
Subsequently, I would no longer have to play around with changing the coordinates (either x, y, or z) in Python.

Paul, what do you think?

Sorry, I just saw your answer, will test that now.



Edited 1 time(s). Last edit at 04/10/2023 03:21PM by Philipp.