Suppr超能文献

用于视网膜眼底图像增强的上下文感知最优传输学习

Context-Aware Optimal Transport Learning for Retinal Fundus Image Enhancement.

作者信息

Vasa Vamsi Krishna, Xiong Yujian, Qiu Peijie, Dumitrascu Oana, Zhu Wenhui, Wang Yalin

机构信息

Arizona State University.

Washington University in St.Louis.

出版信息

IEEE Winter Conf Appl Comput Vis. 2025 Feb-Mar;2025:4016-4025. doi: 10.1109/wacv61041.2025.00395. Epub 2025 Apr 8.

Abstract

Retinal fundus photography offers a non-invasive way to diagnose and monitor a variety of retinal diseases, but is prone to inherent quality glitches arising from systemic imperfections or operator/patient-related factors. However, high-quality retinal images are crucial for carrying out accurate diagnoses and automated analyses. The fundus image enhancement is typically formulated as a distribution alignment problem, by finding a one-to-one mapping between a low-quality image and its high-quality counterpart. This paper proposes a context-informed optimal transport (OT) learning framework for tackling unpaired fundus image enhancement. In contrast to standard generative image enhancement methods, which struggle with handling contextual information (e.g., over-tampered local structures and unwanted artifacts), the proposed context-aware OT learning paradigm better preserves local structures and minimizes unwanted artifacts. Leveraging deep contextual features, we derive the proposed context-aware OT using the earth mover's distance and show that the proposed context-OT has a solid theoretical guarantee. Experimental results on a large-scale dataset demonstrate the superiority of the proposed method over several state-of-the-art supervised and unsupervised methods in terms of signal-to-noise ratio, structural similarity index, as well as two downstream tasks. The code is available at https://github.com/Retinal-Research/Contextual-OT.

摘要

眼底摄影提供了一种非侵入性的方法来诊断和监测各种视网膜疾病,但由于系统缺陷或与操作者/患者相关的因素,它容易出现固有的质量问题。然而,高质量的视网膜图像对于进行准确的诊断和自动分析至关重要。眼底图像增强通常被表述为一个分布对齐问题,即找到低质量图像与其高质量对应图像之间的一一映射。本文提出了一种上下文感知最优传输(OT)学习框架,用于解决未配对的眼底图像增强问题。与标准的生成式图像增强方法不同,后者在处理上下文信息(如过度处理的局部结构和不需要的伪影)方面存在困难,所提出的上下文感知OT学习范式能更好地保留局部结构并减少不需要的伪影。利用深度上下文特征,我们使用推土机距离推导了所提出的上下文感知OT,并表明所提出的上下文OT有坚实的理论保证。在一个大规模数据集上的实验结果表明,所提出的方法在信噪比、结构相似性指数以及两个下游任务方面优于几种最新的监督和无监督方法。代码可在https://github.com/Retinal-Research/Contextual-OT获取。

相似文献

1
Context-Aware Optimal Transport Learning for Retinal Fundus Image Enhancement.用于视网膜眼底图像增强的上下文感知最优传输学习
IEEE Winter Conf Appl Comput Vis. 2025 Feb-Mar;2025:4016-4025. doi: 10.1109/wacv61041.2025.00395. Epub 2025 Apr 8.

引用本文的文献

1
TPOT: TOPOLOGY PRESERVING OPTIMAL TRANSPORT IN RETINAL FUNDUS IMAGE ENHANCEMENT.TPOT:视网膜眼底图像增强中的拓扑保持最优传输
Proc IEEE Int Symp Biomed Imaging. 2025 Apr;2025. doi: 10.1109/isbi60581.2025.10981104. Epub 2025 May 12.

本文引用的文献

5
Optimal Transport for Unsupervised Denoising Learning.用于无监督去噪学习的最优传输
IEEE Trans Pattern Anal Mach Intell. 2023 Feb;45(2):2104-2118. doi: 10.1109/TPAMI.2022.3170155. Epub 2023 Jan 6.
6
An Annotation-Free Restoration Network for Cataractous Fundus Images.无注释白内障眼底图像恢复网络。
IEEE Trans Med Imaging. 2022 Jul;41(7):1699-1710. doi: 10.1109/TMI.2022.3147854. Epub 2022 Jun 30.
7
Modeling and Enhancing Low-Quality Retinal Fundus Images.眼底低质量图像的建模与增强。
IEEE Trans Med Imaging. 2021 Mar;40(3):996-1006. doi: 10.1109/TMI.2020.3043495. Epub 2021 Mar 2.
10
Noise Adaptation Generative Adversarial Network for Medical Image Analysis.用于医学图像分析的噪声适应生成对抗网络。
IEEE Trans Med Imaging. 2020 Apr;39(4):1149-1159. doi: 10.1109/TMI.2019.2944488. Epub 2019 Sep 30.

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验