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.
Downloading models and data from .tar.gz archives instead of .zip (allows extraction via pipe to tar w/o needing to write the entire archive to disk)
Excluding unused FSL binaries
FSL installation is now downloaded from an MSI S3 bucket (was getting frequent timeouts downloading from Oxford mirror)
Add retry on fail to ANTs download
Accomodate CircleCI build testing by excluding host FS attributes from MSI S3-hosted files (i.e. UID/GID, which must be in the range 0-65535 for CircleCI)
Added config.yml for CircleCI build
Updated model directory contents for tasks 514, 515
Reduce disk space needed for Docker build by:
FSL installation is now downloaded from an MSI S3 bucket (was getting frequent timeouts downloading from Oxford mirror)
Add retry on fail to ANTs download
Accomodate CircleCI build testing by excluding host FS attributes from MSI S3-hosted files (i.e. UID/GID, which must be in the range 0-65535 for CircleCI)
Added config.yml for CircleCI build
Updated model directory contents for tasks 514, 515