feta-inference

Inference pipeline for our participation in the FeTA challenge 2021.

Team name: TRABIT

A description of our fetal brain 3D MRI segmentation pipeline can be found in our paper.

Installation

Download the two folders in https://drive.google.com/drive/folders/1V5PBETb89GEA3oSNidTpQRtNADjcdp_0?usp=sharing

Move them to feta-inference/data

Build the docker image by running

cd feta-inference
sh build_docker.sh

The tag for the docker image should be feta_challenge/trabit:latest

Note that you have to rebuild the docker image for changes in the code to be taken into account.

Compute automatic fetal brain MRI segmentation using docker

After you have built the docker image, you can create a docker container and obtain the predicted fetal brain segmentation by running the command

sh example_docker_inference.sh

The script example_docker_inference.sh is based on the instructions found at https://feta-2021.grand-challenge.org/Submission/

You can adapt the script example_docker_inference.sh to segment your fetal brain 3D MRI by changing the following paths at the beginning of the script:

• TEST_INPUT_IMG: path to the folder containing the 3D MRI to be segmented
• TEST_INPUT_META: path to the folder containing meta data about the fetal brain 3D MRI to be segmented
• RESULT_LOCATION: path to the location where to save the output of the segmentation pipeline

The folders TEST_INPUT_IMG and TEST_INPUT_META and the data therein are assumed to use the same structure as for the testing phase of the FeTA challenge 2021.

TEST_INPUT_IMG must contain a folder called \anat containing the 3D MRI to be segmented. The 3D MRI needs to use the file format extension .nii.gz and the naming convention <study-name>_T2w.nii.gz.

In addition, if you already have a brain mask you can put it in the folder TEST_INPUT_IMG with the file name mask.nii.gz. Otherwise, the brain mask will be estimated automatically.

TEST_INPUT_META must contain a file meta.json with the fields Pathology and Gestational age. Pathology can be either Neurotypical or Pathological. Gestational age is the gestational age of the fetus at the time of imaging in weeks.

How to cite

If you find this repository useful in your work, please cite our work

• L. Fidon, M. Aertsen, S. Shit, P. Demaerel, S. Ourselin, J. Deprest, T. Vercauteren. Partial supervision for the FeTA challenge 2021. MICCAI 2021 Perinatal, Preterm and Paediatric Image Analysis (PIPPI) workshop.
• L. Fidon, M. Aertsen, D. Emam, N. Mufti, F. Guffens, T. Deprest, P. Demaerel, A. L. David, A. Melbourne, S. Ourselin, J. Deprest, T. Vercauteren. Label-Set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation MICCAI 2021.

BibTeX:

@article{fidon2021partial,
  title={Partial supervision for the FeTA challenge 2021},
  author={Fidon, Lucas and Aertsen, Michael and Shit, Suprosanna and Demaerel, Philippe and Ourselin, S{\'e}bastien and Deprest, Jan and Vercauteren, Tom},
  journal={arXiv preprint arXiv:2111.02408},
  year={2021}
}
@inproceedings{fidon2021label,
  title={Label-set loss functions for partial supervision: application to fetal brain 3D MRI parcellation},
  author={Fidon, Lucas and Aertsen, Michael and Emam, Doaa and Mufti, Nada and Guffens, Fr{\'e}d{\'e}ric and Deprest, Thomas and Demaerel, Philippe and David, Anna L and Melbourne, Andrew and Ourselin, S{\'e}bastien and others},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={647--657},
  year={2021},
  organization={Springer}
}