Natural Robustness Benchmark for Retinal Vessel Segmentation

Augmented DRIVE, STARE, and CHASE_DB1 vessel-segmentation images for out-of-distribution robustness evaluation.

At a glance

Field	Value
Short name	retinal_vessel_robustness
Full name	Natural Robustness Benchmark for Retinal Vessel Segmentation
Publication date	2024-07-04
Date basis	Initial dataset release
Publication date precision	day
Publication date evidence	zenodo.org/api
Publication date source field	metadata.publication_date (earliest public Zenodo version)
Publication date reviewed	2026-09-11
Date notes	Derivative robustness benchmark release; its own release is used.
Primary category	fundus
Resource role	derivative_dataset
Dataset family	retinal_vessel_robustness
Contained modalities	fundus
Tasks	segmentation
Primary reported quantity	13,024 images
Classes	Not reported (Not reported)
Splits	all
Size	8.15 GB
Source-stated terms	MIT
Normalized terms	mit
Descriptive screening label	Standard label without an explicit NC clause; verify that it applies to data
Terms scope	unknown
Access friction	self_service_authenticated
Route backend	Zenodo
Availability	available (checked 2026-07-21)
Acquisition support	standard_platform_supported
Legacy sample-loader status	Metadata and access only

Reported quantities

Role	Count	Unit	Scope	Basis	Evidence
Primary	13,024	images	Files under the images directories in all three current archives Includes 5,920 DRIVE, 4,144 CHASE, and 2,960 STARE original or augmented images; parent datasets overlap catalog records.	current_deposit_file_listing	zenodo.org/records

Counts retain their source-reported units. Additional rows can describe components, paired items, or derivative copies and are not automatically added to the primary quantity.

Notes

Derivative robustness benchmark from existing vessel datasets.

Documented relationships

These links record source-supported lineage or overlap, not merely similar modality tags.

• This record is derived from chase_db1: The robustness benchmark contains augmented versions of DRIVE, STARE, and CHASE_DB1 images. (evidence)
• This record is derived from drive: The robustness benchmark contains augmented versions of DRIVE, STARE, and CHASE_DB1 images. (evidence)
• This record is derived from stare: The robustness benchmark contains augmented versions of DRIVE, STARE, and CHASE_DB1 images. (evidence)

Access information and download

CLI

# Read-only preflight
eyehub download retinal_vessel_robustness --data-dir ./data --dry-run --json

# Download, only when preflight reports supported behavior
eyehub download retinal_vessel_robustness --data-dir ./data

Python

from eyedatahub.acquisition import preflight_dataset
from eyedatahub.datasets.registry import REGISTRY

ds = REGISTRY.get_dataset('retinal_vessel_robustness')
print(preflight_dataset(ds, './data'))  # no download

Upstream page: zenodo.org/records

Source-term evidence: zenodo.org/records

Loader status

This catalog record provides metadata and access instructions, but it does not yet include a standard DatasetSample loader. Inspect the source file structure or contribute a loader before using it in a training pipeline.

Citation

BibTeX

@misc{retinal_vessel_robustness,
  title  = { Natural Robustness Benchmark for Retinal Vessel Segmentation },
  note   = { Evaluation benchmark for natural robustness evaluation of retinal vessel segmentation models. Zenodo, 2024. doi:10.5281/zenodo.12659652 },
  year   = { 2024 },
  url    = { https://zenodo.org/records/12659652 },
}

Plain text

Evaluation benchmark for natural robustness evaluation of retinal vessel segmentation models. Zenodo, 2024. doi:10.5281/zenodo.12659652

Source-stated terms

• Raw source string: MIT
• Normalized category: mit
• Apparent scope: unknown
• Descriptive screening label: Standard label without an explicit NC clause; verify that it applies to data

⚠️ Source-stated terms, scope, and normalized labels are curation metadata, not legal advice or a permission finding. Review the current official source before transfer or reuse.

Similar resources by shared modality

• airogs: AIROGS: AI for Robust Glaucoma Screening (113,893 images, cc-by-nc-nd)
• multieye: MultiEYE: OCT-Enhanced Fundus Multi-Disease Benchmark (103,959 images, mit)
• eyecare_100k: Eyecare-100K: Multimodal Ophthalmology VQA Corpus (102,000 question answer pairs, unknown)
• justraigs: JustRAIGS: Just Referral AI Glaucoma Screening Dataset (101,442 images, cc-by-nc-nd)
• eyepacs: EyePACS — Diabetic Retinopathy Detection (Kaggle 2015) (88,702 images, research-only)
• angioreport: AngioReport Fundus Angiography Report Dataset (55,361 images, unknown)
• ffa_ir: FFA-IR Medical Report Dataset (47,247 images, unknown)
• mfiddr: MFIDDR: Multi-Field Imaging Dataset for Diabetic Retinopathy (34,452 images, mit)