Effect of Input Resolution on Retinal Vessel Segmentation Performance: An Empirical Study Across Five Datasets

2026-04-03

AI Summary

This study investigates the impact of input image resolution on retinal vessel segmentation performance, with a focus on information loss in fine vessels caused by downsampling. Leveraging five fundus datasets and a U-Net architecture enhanced with Euclidean distance transform to estimate vessel width, the authors propose a vessel-half-width-stratified sensitivity metric. This approach reveals, for the first time, the limitation of the standard Dice score in capturing performance degradation in fine vessel segmentation. Experimental results demonstrate that moderate downsampling of high-resolution images can improve fine vessel sensitivity, whereas low-resolution images perform best at their native resolution. Aggressive downsampling leads to a sensitivity drop of up to 15.8 percentage points for fine vessels, despite negligible changes in Dice score, thereby underscoring the necessity and effectiveness of the proposed evaluation methodology.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

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.

Problem

• retinal vessel segmentation
• input resolution
• thin vessel detection
• downsampling
• evaluation metric

Innovation

• retinal vessel segmentation
• input resolution
• width-stratified sensitivity
• thin vessel detection
• downsampling effect