Unifying Active Learning and Semi-Supervised Learning for Medical Image Segmentation
Quick Answer
This paper shows that The RegAL framework unifies active and semi-supervised learning for medical image segmentation, optimizing annotation selection and unlabeled data usage.
Quick Take
It consistently outperforms state-of-the-art methods across multiple benchmarks, including BraTS 2021 and ProstateX, even with limited labeled data.
Key Points
- RegAL employs topology-aware Pareto optimization for sample acquisition and data utilization.
- It evaluates images based on voxel-wise uncertainty, feature diversity, and topological consistency.
- The framework shows stability with few labeled volumes during training.
- RegAL outperforms existing AL and SSL methods on Dice and boundary-distance metrics.
- Demonstrated effectiveness across datasets like BraTS 2021, dHCP, and ProstateX.
Paper Resources
📖 Reader Mode
~2 min readAbstract:In practical settings, medical image segmentation models are often developed with limited annotated data rather than fully labeled datasets. Training frequently begins in ultra-low labeled regimes where only a small number of volumes are annotated. In such scenarios, practitioners must simultaneously decide which cases to annotate and how to best use the remaining unlabeled data. Although active learning (AL) and semi-supervised learning (SSL) both target annotation scarcity, they are typically designed and optimized independently, resulting in objective mismatch and unstable training during early-stage "cold start" conditions. We propose RegAL, a unified active semi-supervised framework governed by a shared topology-aware Pareto optimization that couples sample acquisition with unlabeled data utilization. RegAL evaluates images along three complementary axes, voxel-wise uncertainty, feature diversity, and a novel topological consistency metric, to select anatomically informative edge cases for annotation. On the other hand, the same criteria are used to identify geometrically stable atlas candidates for diffeomorphic registration-guided augmentation to train a self-supervised Mean Teacher segmentation network. Across BraTS 2021, dHCP, and ProstateX, RegAL remains stable with few labeled volumes and consistently outperforms state-of-the-art AL, SSL, and active semi-supervised baselines across Dice and boundary-distance (ASD, HD95) metrics under extreme annotation scarcity.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML) |
| Cite as: | arXiv:2607.25014 [cs.CV] |
| (or arXiv:2607.25014v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.25014 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Bahram Jafrasteh [view email]
[v1]
Mon, 27 Jul 2026 19:10:12 UTC (10,705 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.