117 color fundus photography images from four public datasets were reorganized and annotated to address the scarcity of precise artery–vein labels. The AVS dataset, created by Wenbo Zhang and colleagues, provides refined vessel segmentation with arterial and venous categories. It was last updated on 2026-05-15.
Use Cases
- Training segmentation models for retinal arteries and veins based on the provided annotations.
- Benchmarking arteriovenous segmentation algorithms on diverse imaging perspectives and pathological conditions.
- Developing diagnostic tools for vascular conditions using the labeled vessel structures.
Strengths
- Contains 117 annotated retinal fundus images.
- Integrates data from four public sources with diverse imaging devices and pathologies.
- Provides refined artery and vein labels beyond basic vessel segmentation.
Limitations
- Row count for annotations is unknown, which may limit suitability assessment.
- Column-level documentation is absent; field semantics must be inferred after download.
Provenance
- Source
- Four publicly available retinal vessel segmentation datasets.
- Collection Method
- Collected, reorganized, and refined from existing public annotations.
- Freshness
- Last updated 2026-05-15 13:30:46.