Entropy-Guided Agreement-Diversity: A Semi-Supervised Active Learning Framework for Fetal Head Segmentation in Ultrasound
Fangyijie Wang, Siteng Ma, Guénolé Silvestre, Kathleen M. Curran
TL;DR
The paper addresses the challenge of scarce labeled fetal ultrasound data for segmentation by integrating active learning with semi-supervised learning. It introduces Entropy-Guided Agreement-Diversity (EGAD), a two-stage AL sampler that first selects uncertain samples via predictive entropy and then refines by an agreement-diversity score combining Cosine similarity and mutual information. The SSL component uses co-training between a lightweight CNN (UNeXt) and a Transformer-based Swin-Unet with a consistency loss and a feature downsampling module to emphasize fetal head regions. On two public datasets with only 5–10% labeled data, SSL-EGAD achieves Dice scores of 94.57% and 96.32%, outperforming existing SSL and AL methods and demonstrating robustness across pregnancy stages.
Abstract
Fetal ultrasound (US) data is often limited due to privacy and regulatory restrictions, posing challenges for training deep learning (DL) models. While semi-supervised learning (SSL) is commonly used for fetal US image analysis, existing SSL methods typically rely on random limited selection, which can lead to suboptimal model performance by overfitting to homogeneous labeled data. To address this, we propose a two-stage Active Learning (AL) sampler, Entropy-Guided Agreement-Diversity (EGAD), for fetal head segmentation. Our method first selects the most uncertain samples using predictive entropy, and then refines the final selection using the agreement-diversity score combining cosine similarity and mutual information. Additionally, our SSL framework employs a consistency learning strategy with feature downsampling to further enhance segmentation performance. In experiments, SSL-EGAD achieves an average Dice score of 94.57\% and 96.32\% on two public datasets for fetal head segmentation, using 5\% and 10\% labeled data for training, respectively. Our method outperforms current SSL models and showcases consistent robustness across diverse pregnancy stage data. The code is available on \href{https://github.com/13204942/Semi-supervised-EGAD}{GitHub}.