Towards Realistic Semi-supervised Medical Image Classification
Wenxue Li, Lie Ju, Feilong Tang, Peng Xia, Xinyu Xiong, Ming Hu, Lei Zhu, Zongyuan Ge
Abstract
Existing semi-supervised learning (SSL) approaches follow the idealized closed-world assumption, neglecting the challenges present in realistic medical scenarios, such as open-set distribution and imbalanced class distribution. Although some methods in natural domains attempt to address the open-set problem, they are insufficient for medical domains, where intertwined challenges like class imbalance and small inter-class lesion discrepancies persist. Thus, this paper presents a novel self-recalibrated semantic training framework, which is tailored for SSL in medical imaging by ingeniously harvesting realistic unlabeled samples. Inspired by the observation that certain open-set samples share some similar disease-related representations with in-distribution samples, we first propose an informative sample selection strategy that identifies high-value samples to serve as augmentations, thereby effectively enriching the semantics of known categories. Furthermore, we adopt a compact semantic clustering strategy to address the semantic confusion raised by the above newly introduced open-set semantics. Moreover, to mitigate the interference of class imbalance in open-set SSL, we introduce a less biased dual-balanced classifier with similarity pseudo-label regularization and category-customized regularization. Extensive experiments on a variety of medical image datasets demonstrate the superior performance of our proposed method over state-of-the-art Closed-set and Open-set SSL methods.
BibTeX
@article{Li_Ju_Tang_Xia_Xiong_Hu_Zhu_Ge_2025, title={Towards Realistic Semi-supervised Medical Image Classification}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/32526}, DOI={10.1609/aaai.v39i5.32526}, abstractNote={Existing semi-supervised learning (SSL) approaches follow the idealized closed-world assumption, neglecting the challenges present in realistic medical scenarios, such as open-set distribution and imbalanced class distribution. Although some methods in natural domains attempt to address the open-set problem, they are insufficient for medical domains, where intertwined challenges like class imbalance and small inter-class lesion discrepancies persist. Thus, this paper presents a novel self-recalibrated semantic training framework, which is tailored for SSL in medical imaging by ingeniously harvesting realistic unlabeled samples. Inspired by the observation that certain open-set samples share some similar disease-related representations with in-distribution samples, we first propose an informative sample selection strategy that identifies high-value samples to serve as augmentations, thereby effectively enriching the semantics of known categories. Furthermore, we adopt a compact semantic clustering strategy to address the semantic confusion raised by the above newly introduced open-set semantics. Moreover, to mitigate the interference of class imbalance in open-set SSL, we introduce a less biased dual-balanced classifier with similarity pseudo-label regularization and category-customized regularization. Extensive experiments on a variety of medical image datasets demonstrate the superior performance of our proposed method over state-of-the-art Closed-set and Open-set SSL methods.}, number={5}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Li, Wenxue and Ju, Lie and Tang, Feilong and Xia, Peng and Xiong, Xinyu and Hu, Ming and Zhu, Lei and Ge, Zongyuan}, year={2025}, month={Apr.}, pages={4968-4976} }