There is a newer version of the record available.

Published July 30, 2021 | Version v2
Dataset Restricted

Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation

  • 1. School of Biomedical Engineering & Imaging Sciences, King's College London, UK
  • 2. Department of Radiology, University Hospitals Leuven, Belgium
  • 3. Department of Obstetrics and Gynaecology, University Hospitals Leuven, Belgium
  • 4. Institute for Women's Health, University College London, UK

Description

Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation

This repository contains data contributed and used in our MICCAI 2021 paper:
Fidon, L. et al. Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation.

Our code is publicly available at
https://github.com/LucasFidon/fetal-brain-segmentation-partial-supervision-miccai21

The FeTA 3D MRIs and segmentations contained in this repository shall be used only for research and education purposes.

Updates:

Version 2 (Avril 2022)

3D MRI and refined manual segmentation have been added for:
- the 10 fetal brain MRIs of the testing set of the FeTA data release 1 (sub-feta081 to sub-feta090).
- the 40 fetal brain MRIs new in the FeTA data release 2 (sub-041 to sub080).
The same pre-processing as in version 1 below has been performed.
The code used for the pre-processing is available at
https://github.com/LucasFidon/fetal-seg-preprocessing

Some additional refinements of the manual segmentation have been performed for sub-feta001 to sub-feta040.

See participant.tsv for more information about the 3D MRI studies.

The pathology for each study has been diagnosed by two radiologists in our team 
and can be found in participant.tsv.
Those are not present in the original FeTA dataset release 2.1

How to cite:

If you use the deep neural network weights please cite:

  • Fidon, L., Aertsen, M., Emam, D., et al. Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation. MICCAI (2021).

if you use the FeTA data (release 1 and 2) and the corrected segmentations that we contributed please cite:

  • Payette, K., de Dumast, P., Kebiri, H. et al. An automatic multi-tissue human fetal brain segmentation benchmark using the Fetal Tissue Annotation Dataset. Sci Data 8, 167 (2021). https://doi.org/10.1038/s41597-021-00946-3
  • Fidon, L., Aertsen, M., Emam, D., et al. Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation. MICCAI (2021).

Terms of use - FeTA 3D MRIs and segmentations:

This is an agreement (“Agreement”) between you the downloader (“Downloader”) and the owner of the materials (“User”) governing the use of the Fetal Tissue Annotation and Segmentation Dataset to be downloaded.

I. Acceptance of this Agreement

By downloading or otherwise accessing the Fetal Tissue Annotation and Segmentation Dataset, the Downloader represents his/her acceptance of the terms of this Agreement.

II. Data ownership

The owner of the Fetal Tissue Annotation and Segmentation Dataset is the University Children’s Hospital Zurich.

III. Use of the Materials

Fetal Tissue Annotation and Segmentation Dataset is used only for research and education. Any other kind of use you will lead to the recall of all datasets, stop of collaboration, and legal consequences. This Agreement represents the entire agreement between Downloader and User with respect to the downloading and use of the Materials and supersedes all prior or contemporaneous communications and proposals (whether oral, written, or electronic) between Downloader and User with respect to downloading or using the Materials. 

 

Files

Restricted

The record is publicly accessible, but files are restricted to users with access.

Request access

If you would like to request access to these files, please fill out the form below.

You need to satisfy these conditions in order for this request to be accepted:

The files contained in this repository shall be used only for research and education purposes.
Access will be granted to anyone satisfying this condition.

You are currently not logged in. Do you have an account? Log in here

Additional details

Related works

Is supplement to
Conference paper: https://arxiv.org/abs/2107.03846 (URL)

Funding

TRABIT – Translational Brain Imaging Training Network 765148
European Commission