提出统一模型，让单一网络跨多种眼底图像分割血管。
Convolutional Prompting for Broad-Domain Retinal Vessel Segmentation
- 用双维度局部提示提取不同图像域的特征。
- 在5种数据集上优于8个基线方法，平均DICE达0.912。
- 无需改动主网络结构，适配各类眼科影像分析任务。
以往视网膜血管分割研究集中于特定图像域，主要针对彩色眼底照相（CFP）。本文首次挑战更复杂的广域视网膜血管分割（BD-RVS）任务，旨在构建一个适用于CFP、SLO、UWF、OCTA和FFA等多种图像域的统一模型。为此，提出双卷积提示（DCP）模块，通过在空间位置与通道维度上进行局部提示，学习域特异性特征。DCP作为即插即用模块，可有效将基于R2AU-Net的分割网络升级为统一模型，无需修改原网络结构。为评估，我们整合了五个公开的域专用数据集（ROSSA、FIVES、IOSTAR、PRIME-FP20和VAMPIRE），构建广域数据集。为基准测试，重新应用多个原有方法，共生成8个基线模型。大量实验表明，所提方法在广域数据集上显著优于各基线，在五域平均DICE达0.912。
原文摘要 · Abstract (English)
Previous research on retinal vessel segmentation is targeted at a specific image domain, mostly color fundus photography (CFP). In this paper we make a brave attempt to attack a more challenging task of broad-domain retinal vessel segmentation (BD-RVS), which is to develop a unified model applicable to varied domains including CFP, SLO, UWF, OCTA and FFA. To that end, we propose Dual Convoltuional Prompting (DCP) that learns to extract domain-specific features by localized prompting along both position and channel dimensions. DCP is designed as a plug-in module that can effectively turn a R2AU-Net based vessel segmentation network to a unified model, yet without the need of modifying its network structure. For evaluation we build a broad-domain set using five public domain-specific datasets including ROSSA, FIVES, IOSTAR, PRIME-FP20 and VAMPIRE. In order to benchmark BD-RVS on the broad-domain dataset, we re-purpose a number of existing methods originally developed in other contexts, producing eight baseline methods in total. Extensive experiments show the the proposed method compares favorably against the baselines for BD-RVS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。