I'm guessing that this thread is moving along the same scope as one of your other threads, where you're trying to use afni_proc.py to do certain things, pause and do registration in FNIRT and then continue on the afni_proc.py script.

If the registration via FNIRT is the only thing happening outside of AFNI tools, one thing you could try is to allow afni_proc.py to generate the entire script for you and then use an empty block in the -blocks command.

-blocks tshift volreg empty blur scale regress

Doing so may allow you to use the working afni_proc.py code with minimal modifications.

The other thing that I will mention is that if you're having poor performance in 3dQwarp due to Skull Stripping issues, you could always remove the skull before the afni_proc.py and pass "-anat_has_skull no" to the script. I've found that 3dSkullStrip can do a decent job, but occasionally (depending on image characteristics) FSL's VBM script or Freesurfer may work better for your particular data.



Edited 3 time(s). Last edit at 07/22/2014 11:52AM by Peter Molfese.