AFNI Message Board

History of AFNI updates  

|
June 14, 2021 06:34PM
Hello,

I have a simple monkey/animal_warper/afni_proc.py question. I am modifying the MACAQUE_DEMO_REST pipeline.
If I want to use eroded WM signal as a nuisance regressor in afni_proc (whether via anaticor or as averaged WM), and I have WM mask from the NMT template, I should import the mask in NMT space into afni_proc, correct?

So this:

afni_proc.py                                                              \
[...]
        -mask_import             WM_mask ${additional_path}/WMe_mask+tlrc     \
[...]
        -regress_ROI             WM_mask                                      \

or this

afni_proc.py                                                              \
[...]
        -mask_import             WM_mask ${additional_path}/WMe_mask+tlrc     \
[...]
        -regress_anaticor_fast                                                \
        -regress_anaticor_radius  20                                          \
        -regress_anaticor_label   WM_mask                                      \

instead of
afni_proc.py                                                              \
[...]
        -anat_follower_ROI       WMmask epi ${odir_aw}'/'WM_anat_mask+orig    \
        -anat_follower_erode     WMmask                                       \
[...]                       
        -regress_anaticor_fast                                                \
        -regress_anaticor_radius  20                                          \
        -regress_anaticor_label   WMmask                                      \

because the latter would involve warping the WM mask from the NMT to individual space in animal warper and then warping it back to NMT space in afni_proc, right?
Subject Author Posted

regressing out WM in afni_proc.py

Pawel June 14, 2021 06:34PM

Re: regressing out WM in afni_proc.py

rick reynolds July 15, 2021 10:47AM

Re: regressing out WM in afni_proc.py

Pawel July 15, 2021 10:52AM



Sorry, only registered users may post in this forum.

Click here to login