SUIT Cerebellum Probabilistic Mask Segmentation

[JW1]

Dear AFNI experts,

I am interested in extracting cerebellum segmentation mask from SUIT cerebellum probabilistic mask (attached) and use that segmentation for the seed for resting-state analysis. Could you please let me know if there is any way to extract the each component mask?

Best,
JW

[dglen]

You have already downloaded the atlas dataset from our website, so first change to that directory

cd dir_cerebellum # use the directory where you put it

set AFNI environment variable to use the new atlases in this directory

@AfniEnv -set AFNI_SUPP_ATLAS_DIR pwd

check if the SUIT cerebellum atlases are recognized by AFNI

whereami -show_atlases

see the list of atlas structures in the header of one of the atlases

3dinfo -atlas_points Cerebellum-SUIT.nii.gz

extract one of the regions from that atlas - using 3dcalc or any AFNI C program

3dcalc -a “Cerebellum-SUIT::Left_Fastigial” -expr a -prefix leftfast.nii.gz

another way to extract a region - using whereami

whereami -prefix rightfast.nii.gz -mask_atlas_region “Cerebellum-SUIT::Right_Fastigial”

resample to EPI grid (grid of some dataset in MNI space) starting with the atlas again

3dresample -master ~/AFNI_data6/afni/epi_r1_MNI+tlrc. -inset “Cerebellum-SUIT::Right_Fastigial” -prefix right_fast_rs.nii.gz

[JW1]

Hi Daniel,

Thank you so much for the instruction. Could you please clarify one thing? I am a little bit confused by the last step (resample to EPI grid (grid of some dataset in MNI space) starting with the atlas again) for the two reasons.

First, I could not find epi_r1_MNI+tlrc in the directory you suggested. I could only find epi_r1+orig. Did I make a mistake when installing AFNI or can I just simply convert the orig to tlrc?

Second, regarding the 3dresample, I don’t need to 3dresample the extracted mask to errts.fanaticor in the proc results data of the participants? Can I just simply use the extracted mask as a seed for the connectivity analysis?

Thank you in advance for your help!

Best,
JW

[dglen]

Those are just examples of different ways to use masks. If you have your own EPI dataset, you can use a command like that one to resample the mask to your EPI dataset that is in the same space as the atlas.

[JW1]

Hi Daniel,

Thanks for the clarification! That really helps.

Best,
Jun