unravel.segment.brain_mask module

Use seg_brain_mask (sbm) from UNRAVEL to use a trained ilastik project (pixel classification) to mask the brain in resampled autofluo images (often improves registration).

Prereqs:

• Organize training tif slices (from seg_copy_tifs) into a single folder.

• Train an Ilastik project to segment the brain (https://b-heifets.github.io/UNRAVEL/guide.html#train-an-ilastik-project).

• Train ilastik (tissue = label 1) w/ tifs from reg_inputs/autofl_`*`um_tifs/*.tif (from reg_prep)

• Save brain_mask.ilp in experiment directory of use -ilp

Inputs:

• reg_inputs/autofl_`*`um.nii.gz

• brain_mask.ilp # in exp dir

Outputs:

• reg_inputs/autofl_`*`um_tifs_ilastik_brain_seg/slice_`*`.tif series

• reg_inputs/autofl_`*`um_brain_mask.nii.gz (can be used for reg and vstats_z_score)

• reg_inputs/autofl_`*`um_masked.nii.gz

Note

• Ilastik executable files for each OS (update path and version as needed):

• Linux and WSL: /usr/local/ilastik-1.4.0.post1-Linux/run_ilastik.sh

• Mac: /Applications/ilastik-1.4.0.post1-OSX.app/Contents/ilastik-release/run_ilastik.sh

• Windows: C:Program Filesilastik-1.4.0.post1run_ilastik.bat

• The Ilastik project should be closed before running this script.

Next command:

• reg

Usage:

seg_brain_mask -ilp <path/brain_mask.ilp> [-ie path/ilastik_executable ] [-i reg_inputs/autofl_50um.nii.gz] [-d list of paths] [-p sample??] [-v]

unravel.segment.brain_mask.parse_args()

unravel.segment.brain_mask.main()