Uncertainty Global Contrastive Learning Framework for Semi-Supervised Medical Image Segmentation
- 6 November 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 29 (1) , 433-442
Abstract
In semi-supervised medical image segmentation, the issue of fuzzy boundaries for segmented objects arises. With limited labeled data and the interaction of boundaries from different segmented objects, classifying segmentation boundaries becomes challenging. To mitigate this issue, we propose an uncertainty global contrastive learning (UGCL) framework. Specifically, we propose a patch filtering method and a classification entropy filtering method to provide reliable pseudo-labels for unlabelled data, while separating fuzzy boundaries and high-entropy pixel points as unreliable points. Considering that unreliable regions contain rich complementary information, we introduce an uncertainty global contrast learning method to distinguish these challenging unreliable regions, enhancing intra-class compactness and inter-class separability at the global data level. Within our optimization framework, we also integrate consistency regularization techniques and select unreliable points as targets for consistency. As demonstrated, the contrastive learning and consistency regularization applied to uncertain points enable us to glean valuable semantic information from unreliable data, which enhances segmentation accuracy. We evaluate our method on two publicly available medical image datasets and compare it with other state-of-the-art semi-supervised medical image segmentation methods, and a series of experimental results show that our method has achieved substantial improvements.Keywords
Funding Information
- National Natural Science Foundation of China (41804112)
This publication has 30 references indexed in Scilit:
- DiffRect: Latent Diffusion Label Rectification for Semi-supervised Medical Image Segmentation2024
- Cross Prompting Consistency with Segment Anything Model for Semi-supervised Medical Image Segmentation2024
- Cross-Adversarial Local Distribution Regularization for Semi-supervised Medical Image Segmentation2023
- Ambiguity-selective consistency regularization for mean-teacher semi-supervised medical image segmentationMedical Image Analysis, 2023
- Exploring Smoothness and Class-Separation for Semi-supervised Medical Image Segmentation2022
- Distributed contrastive learning for medical image segmentationMedical Image Analysis, 2022
- Medical image segmentation using deep learning: A surveyIET Image Processing, 2022
- A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imagingMedical Image Analysis, 2021
- Uncertainty-Aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentation2019
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015