arXiv 2509.236172025-09-28cs.CVcs.AI

BioVessel-Net 与 RetinaMix：基于 OCTA 图像的无监督视网膜血管分割

BioVessel-Net and RetinaMix: Unsupervised Retinal Vessel Segmentation from OCTA Images

University of Electronic Science and Technology of China（电子科技大学）；Southern Methodist University（南方法律大学）；University of Texas Southwestern Medical Center（德克萨斯西南医学中心）；University Medical Center Utrecht（乌得勒支大学医学中心）；Georgetown University（杰斐逊大学）；Sichuan Provincial People’s Hospital（四川省级人民医院）；University of Connecticut（康涅狄格大学）

Cheng Huang, Weizheng Xie, Fan Gao, Yutong Liu, Ruoling Wu, Zeyu Han, Jingxi Qiu, Xiangxiang Wang, Zhenglin Yang, Hao Wang, Yongbin Yu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出融合血管生物统计、对抗细化与半径引导分割的无监督框架 BioVessel-Net，并构建 RetinaMix 基准，实现无需标注的高精度可解释 OCTA 视网膜血管分割。

摘要：

视网膜血管的结构变化是青光眼及其他眼部疾病发生和进展的关键生物标志物。然而，当前的血管分割方法在很大程度上依赖监督学习和大量人工标注，而这些标注成本高昂、容易出错，并且在光学相干断层扫描血管成像中难以获得。本文提出 BioVessel-Net，这是一种无监督生成框架，融合了血管生物统计学、对抗式细化和半径引导的分割策略。与基于像素的方法不同，BioVessel-Net 直接对具有生物统计一致性的血管结构进行建模，从而在无需标注数据或高性能计算的情况下实现准确且可解释的血管提取。为支持训练和评估，我们引入 RetinaMix，这是一个新的基准数据集，包含来自不同人群、具有高分辨率血管细节的二维和三维 OCTA 图像。实验结果表明，BioVessel-Net 在 RetinaMix 和现有数据集上均取得近乎完美的分割精度，显著优于最先进的监督和半监督方法。BioVessel-Net 与 RetinaMix 共同为视网膜血管分析提供了一种无标签、计算高效且临床可解释的解决方案，在青光眼监测、血流建模和进展预测方面具有广泛潜力。代码和数据集可在 https://github.com/VikiXie/SatMar8 获取。

英文摘要：

Structural changes in retinal blood vessels are critical biomarkers for the onset and progression of glaucoma and other ocular diseases. However, current vessel segmentation approaches largely rely on supervised learning and extensive manual annotations, which are costly, error-prone, and difficult to obtain in optical coherence tomography angiography. Here we present BioVessel-Net, an unsupervised generative framework that integrates vessel biostatistics with adversarial refinement and a radius-guided segmentation strategy. Unlike pixel-based methods, BioVessel-Net directly models vascular structures with biostatistical coherence, achieving accurate and explainable vessel extraction without labeled data or high-performance computing. To support training and evaluation, we introduce RetinaMix, a new benchmark dataset of 2D and 3D OCTA images with high-resolution vessel details from diverse populations. Experimental results demonstrate that BioVessel-Net achieves near-perfect segmentation accuracy across RetinaMix and existing datasets, substantially outperforming state-of-the-art supervised and semi-supervised methods. Together, BioVessel-Net and RetinaMix provide a label-free, computationally efficient, and clinically interpretable solution for retinal vessel analysis, with broad potential for glaucoma monitoring, blood flow modeling, and progression prediction. Code and dataset are available: https://github.com/VikiXie/SatMar8.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。