Quality-Guided Semi-Supervised Learning for Medical Image Segmentation
2026-06-01 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the challenge of training medical image segmentation models when only a small amount of labeled data is available by using semi-supervised learning. Instead of relying on the model's own confidence to judge the quality of predictions for unlabeled data, they train a separate network to estimate how good a segmentation is. This quality predictor helps improve learning by guiding regularization and weighting pseudo-labeled samples based on their estimated quality. Their experiments show that this approach improves performance across different datasets and models, making semi-supervised segmentation more effective.
medical image segmentationsemi-supervised learningpseudo-labelingquality estimationregularizationunlabeled datadeep learningmodel confidenceimage-mask pairssample reweighting
Authors
Kumar Abhishek, Ghassan Hamarneh
Abstract
Training accurate medical image segmentation models requires large amounts of densely annotated data, which is costly and time-consuming to obtain. Semi-supervised learning (SSL) alleviates this by learning from both abundant unlabeled data and limited labeled data. However, most modern SSL methods rely on pseudolabels for unlabeled data, and typically assess their reliability through model confidence or uncertainty, measures that are self-referential and lack explicit grounding in segmentation quality. Instead, we propose a quality-guided SSL framework that trains a dedicated network to estimate segmentation quality from image-mask pairs. The predictor is trained on variable-quality masks generated through synthetic corruptions augmented with imperfect outputs from partially trained segmentation models, capturing realistic error patterns encountered during training. We integrate the quality predictor into SSL through two complementary mechanisms: a quality-aware regularization loss and a quality-based pseudolabel sample reweighting scheme. We show that our method serves as a drop-in enhancement to existing SSL frameworks. Extensive experiments across five datasets and multiple architectures demonstrate consistent improvements over competing SSL methods, advancing the state-of-the-art in semi-supervised medical image segmentation.