Partial supervision for the FeTA challenge 2021

TL;DR

This paper presents a method using label-set loss functions for partially supervised learning to improve fetal brain MRI segmentation by merging multiple datasets with different annotation protocols, enhancing generalizability without extra hyper-parameter tuning.

Contribution

It introduces the application of label-set loss functions to combine heterogeneous datasets for fetal brain MRI segmentation, addressing annotation inconsistencies.

Findings

Improved segmentation performance on fetal brain MRI datasets.

Effective merging of datasets with different annotation protocols.

No additional hyper-parameter tuning required.

Abstract

This paper describes our method for our participation in the FeTA challenge2021 (team name: TRABIT). The performance of convolutional neural networks for medical image segmentation is thought to correlate positively with the number of training data. The FeTA challenge does not restrict participants to using only the provided training data but also allows for using other publicly available sources. Yet, open access fetal brain data remains limited. An advantageous strategy could thus be to expand the training data to cover broader perinatal brain imaging sources. Perinatal brain MRIs, other than the FeTA challenge data, that are currently publicly available, span normal and pathological fetal atlases as well as neonatal scans. However, perinatal brain MRIs segmented in different datasets typically come with different annotation protocols. This makes it challenging to combine those datasets to train a deep neural network. We recently proposed a family of loss functions, the label-set loss functions, for partially supervised learning. Label-set loss functions allow to train deep neural networks with partially segmented images, i.e. segmentations in which some classes may be grouped into super-classes. We propose to use label-set loss functions to improve the segmentation performance of a state-of-the-art deep learning pipeline for multi-class fetal brain segmentation by merging several publicly available datasets. To promote generalisability, our approach does not introduce any additional hyper-parameters tuning.