DCAN-Labs / BIBSnet

This BIDS App provides the utility of creating a nnU-Net anatomical MRI segmentation and mask with a infant brain trained model. It can easily be included in other processing pipelines and for circumventing JLF within Nibabies.
https://bibsnet.readthedocs.io/en/latest/
Apache License 2.0
4 stars 8 forks source link

postBIBSnet chirality correction not accurate around midline #84

Closed moser297 closed 2 months ago

moser297 commented 1 year ago

What happened?

Chirality correction during postBIBSnet is in many cases not accurate around the medial plane (see example image) which causes the MCRIBS surface reconstruction in Nibabies to crash.

image

What command did you use?

# run cabinet 
env -i ${singularity} run \
-B ${data_dir}/:/input \
-B ${out_dir}/:/output \
-B /home/faird/shared/code/external/utilities/freesurfer_license/license.txt:/opt/freesurfer/license.txt \
/home/faird/shared/code/internal/pipelines/cabinet_container/cabinet_shrink_docker_build_plus_clusters_fix_09272023a.sif \
/input /output participant -jargs /home/cabinet/parameter-file-container.json -start prebibsnet -end postbibsnet -v

What version of BIBSnet are you using?

latest

Relevant log output

No response

Add any additional information or context about the problem here.

Happy to help solving this issue as I have plenty of datasets where I'm planning to use the Nibabies MCRIBS workflow combined with BIBSnet.

LuciMoore commented 1 year ago

@moser297 this error should be largely ameliorated in the latest version of BIBSNet that includes a corrected 1mo chirality template. @tikal004 is working on incorporating corrected templates for the older ages as well currently

Can you please re-run your data and let us know how they compare?

kellyhi08 commented 9 months ago

Hello, I am running into a similar problem, where the left/right labeling is going awry.

Command I am using (I downloaded the latest bibsnet version on Jan 26th): singularity pull bibsnet.sif docker://dcanumn/bibsnet:latest singularity run --nv --cleanenv --no-home \ -B $path/BIBSnet_inputs/:/input \ -B $path/BIBSnet_outputs:/output \ $path/BIBSnet/bibsnet.sif \ /input /output participant --participant-label 001

Any updates or suggestions for fixing this?

kellyhi08 commented 9 months ago

Image: (subject age= 11 months)

image
LuciMoore commented 2 months ago

Resolved in latest version that no longer produces nnU-Net outputs with chirality errors https://github.com/DCAN-Labs/BIBSnet/releases/tag/3.4.1