Unsupervised Segmentation of Fetal Brain MRI using Deep Learning Cascaded Registration

TL;DR

This paper introduces an unsupervised fetal brain MRI segmentation method using cascaded deep learning registration, eliminating the need for extensive labeled training data and achieving competitive performance.

Contribution

It presents a novel cascaded deep learning registration approach for multi-atlas segmentation that requires no training labels, improving accuracy and efficiency in fetal brain MRI analysis.

Findings

Outperforms state-of-the-art registration methods.

Achieves similar segmentation performance to nnU-Net with less annotated data.

Operates with comparable inference time to existing methods.

Abstract

Accurate segmentation of fetal brain magnetic resonance images is crucial for analyzing fetal brain development and detecting potential neurodevelopmental abnormalities. Traditional deep learning-based automatic segmentation, although effective, requires extensive training data with ground-truth labels, typically produced by clinicians through a time-consuming annotation process. To overcome this challenge, we propose a novel unsupervised segmentation method based on multi-atlas segmentation, that accurately segments multiple tissues without relying on labeled data for training. Our method employs a cascaded deep learning network for 3D image registration, which computes small, incremental deformations to the moving image to align it precisely with the fixed image. This cascaded network can then be used to register multiple annotated images with the image to be segmented, and combine the propagated labels to form a refined segmentation. Our experiments demonstrate that the proposed cascaded architecture outperforms the state-of-the-art registration methods that were tested. Furthermore, the derived segmentation method achieves similar performance and inference time to nnU-Net while only using a small subset of annotated data for the multi-atlas segmentation task and none for training the network. Our pipeline for registration and multi-atlas segmentation is publicly available at https://github.com/ValBcn/CasReg.