研究了图像分辨率对视网膜血管分割的影响，发现低分辨率会严重损失细血管信息。
Effect of Input Resolution on Retinal Vessel Segmentation Performance: An Empirical Study Across Five Datasets
- 提出按血管宽度分层的敏感性评估方法，区分细、中、粗血管检测效果
- 高分辨率数据集在下采样至256-876像素时细血管敏感性提升，低分辨率数据集则以原分辨率最佳
- 标准Dice分数无法反映细血管信息丢失，最大敏感性下降达15.8个百分点
大多数深度学习视网膜血管分割流程为满足GPU内存限制和统一批量处理，将眼底图像进行缩放。然而，这种缩放对细血管检测的影响尚未深入研究。高分辨率图像下采样后，细血管退化为亚像素结构，造成不可逆的信息损失。标准体积指标如Dice分数无法捕捉该损失，因粗血管像素主导评价结果。本研究在五个眼底数据集（DRIVE、STARE、CHASE_DB1、HRF、FIVES）上，固定其他设置，训练基础UNet模型于多种下采样比例。引入基于宽度分层的敏感性指标，通过欧氏距离变换获取原始分辨率下的血管宽度估计，分别评估半宽小于3像素（细）、3至7像素（中）、大于7像素（粗）的血管检测表现。结果显示：在高分辨率数据集（HRF、FIVES）中，细血管敏感性随图像下采样至编码器有效工作范围而单调提升，峰值出现在处理宽度256至876像素之间；而在低至中等分辨率数据集（DRIVE、STARE、CHASE_DB1）中，细血管敏感性在或接近原分辨率时最高，任何下采样均导致性能下降。所有五组数据中，激进下采样使细血管敏感性最多下降15.8个百分点（DRIVE），而Dice分数相对稳定，证实仅依赖Dice评估微血管分割存在缺陷。
原文摘要 · Abstract (English)
Most deep learning pipelines for retinal vessel segmentation resize fundus images to satisfy GPU memory constraints and enable uniform batch processing. However, the impact of this resizing on thin vessel detection remains underexplored. When high resolution images are downsampled, thin vessels are reduced to subpixel structures, causing irreversible information loss even before the data enters the network. Standard volumetric metrics such as the Dice score do not capture this loss because thick vessel pixels dominate the evaluation. We investigated this effect by training a baseline UNet at multiple downsampling ratios across five fundus datasets (DRIVE, STARE, CHASE_DB1, HRF, and FIVES) with native widths ranging from 565 to 3504 pixels, keeping all other settings fixed. We introduce a width-stratified sensitivity metric that evaluates thin (half-width <3 pixels), medium (3 to 7 pixels), and thick (>7 pixels) vessel detection separately, using native resolution width estimates derived from a Euclidean distance transform. Results show that for high-resolution datasets (HRF, FIVES), thin vessel sensitivity improves monotonically as images are downsampled toward the encoder's effective operating range, peaking at processed widths between 256 and 876 pixels. For low-to-mid resolution datasets (DRIVE, STARE, CHASE_DB1), thin vessel sensitivity is highest at or near native resolution and degrades with any downsampling. Across all five datasets, aggressive downsampling reduced thin vessel sensitivity by up to 15.8 percentage points (DRIVE) while Dice remained relatively stable, confirming that Dice alone is insufficient for evaluating microvascular segmentation.
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。