Exploring Test Time Adaptation for Subcortical Segmentation of the Fetal Brain in 3D Ultrasound

2025-02-12

AI Summary

To address the significant performance degradation of automatic subcortical structure segmentation in fetal 3D ultrasound under cross-device and cross-operator domain shifts, this paper proposes an anatomy-constrained test-time adaptation (TTA) method. Our approach uniquely integrates normative atlas priors into the TTA framework for the first time, enforcing anatomical consistency via an atlas-guided loss function that steers gradient updates during inference. To further enhance robustness, we incorporate multimodal domain shift simulation during adaptation. Evaluated on both real-world and synthetically generated domain-shifted datasets, our method achieves state-of-the-art performance, improving Dice scores by 3.2–5.8% over existing approaches. The implementation is publicly available, offering a generalizable and interpretable clinical solution for automated fetal brain development monitoring.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Segmentation
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Monitoring the growth of subcortical regions of the fetal brain in ultrasound (US) images can help identify the presence of abnormal development. Manually segmenting these regions is a challenging task, but recent work has shown that it can be automated using deep learning. However, applying pretrained models to unseen freehand US volumes often leads to a degradation of performance due to the vast differences in acquisition and alignment. In this work, we first demonstrate that test time adaptation (TTA) can be used to improve model performance in the presence of both real and simulated domain shifts. We further propose a novel TTA method by incorporating a normative atlas as a prior for anatomy. In the presence of various types of domain shifts, we benchmark the performance of different TTA methods and demonstrate the improvements brought by our proposed approach, which may further facilitate automated monitoring of fetal brain development. Our code is available at https://github.com/joshuaomolegan/TTA-for-3D-Fetal-Subcortical-Segmentation.

Problem

• Automate fetal brain subcortical segmentation in 3D ultrasound.
• Address performance degradation in unseen freehand US volumes.
• Improve model adaptation using test time adaptation methods.

Innovation

• Test Time Adaptation
• Normative atlas prior
• Automated fetal monitoring