Document Type
Article
Publication Date
10-1-2025
Abstract
Contrast-enhanced knowledge distillation promises to transform medical diagnostics and reveal promising approaches for tumor segmentation on non-contrast medical images. However, existing methods related to contrast-enhanced knowledge distillation still make it hard to distill reliable contrast-enhanced knowledge for tumor segmentation due to the limitations of (1) unable to quantify uncertainty information for reliable contrast-enhanced and non-contrast knowledge modeling, which leads to an over-confidence cross-domain adaptation for transferring contrast-enhanced knowledge; (2) using vision information only ignores rich semantic features in medical language, which make it hard to model complex tumor enhancement feature. In this study, we propose an evidence-guided and tumor-aware knowledge distillation (EGTA-KD) for transferring contrast-enhanced domain knowledge to non-contrast domain knowledge. Specifically, to achieve tumor-awareness by embedding semantic features from text, the tumor-aware cross-modal synchronizer (TACMS) is proposed to calculate tumor score maps for matching pixel wise image and text features. To achieve reliable cross-domain modeling for transferring contrast-enhanced knowledge, the innovative uncertainty-quantified evidence unit (UQEU) parameterizes the probability distribution within subjective logic to gather reliable evidence of contrast-enhanced knowledge while quantifying the uncertainty of prediction. Lastly, newly designed dual-level knowledge distillation (DLKD) minimizes tumor score map errors and matches evidence distribution for uncertainty-aware contrast-enhanced knowledge distillation. Extensive experiments of tumor segmentation on non-contrast medical images are performed using multi-modality medical image datasets (i.e., Brain MRI dataset, Liver MRI dataset, and Kidney CT dataset). Experimental results demonstrate the proposed EGTA-KD outperforms the other compared state-of-the-art methods, revealing its superiority of tumor segmentation on non-contrast medical images via uncertainty-aware contrast-enhanced knowledge distillation.
Keywords
evidence learning, knowledge distillation, segmentation, uncertainty
Language
English
Publication Title
Medical Image Analysis
Rights
© 2025 The Authors. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/), which permits non-commercial copying and redistribution of the material in any medium or format, provided the original work is not changed in any way and is properly cited.
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Recommended Citation
Zhao, J., & Li, S. (2025). When evidence modeling meets knowledge distillation: Towards reliable contrast-enhanced knowledge distillation for non-contrast medical image segmentation. Medical Image Analysis, 105, 103677. https://doi.org/10.1016/j.media.2025.103677
Manuscript Version
Final Publisher Version