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.
@tjhendrickson you can merge once the new T1- and T2-only models are ported into the container. It's tested and should work (I also tested T1- and T2-only just using the older models)
Updates include:
updated version to v3.3.0
removed chirality correction from postbibsnet
removed additional code/data no longer needed with CC removed, including bin folder (that contained LR mask registration bash script) in container and chirality templates folder in data.tar.gz on s3
in tar.gz file on s3, also removed all models except 540 (because the older models shouldn't be run without chirality correction anyway) and updated the models.csv file under data.tar.gz
@tjhendrickson you can merge once the new T1- and T2-only models are ported into the container. It's tested and should work (I also tested T1- and T2-only just using the older models) Updates include: