DeepDRiD: Diabetic Retinopathy Grading and Image Quality Dataset

2,000 regular fundus images (500 patients × 2 fields × 2 eyes) plus 256 ultra-widefield fundus images labeled for DR severity (ICDR grades 0-4) and image quality assessment (gradable/ungradable). From the DeepDRiD challenge (MICCAI 2020 / ISBI 2020).

At a glance

Field	Value
Short name	`deepdrid`
Full name	DeepDRiD: Diabetic Retinopathy Grading and Image Quality Dataset
Primary category	`fundus`
Contained modalities	fundus, uwf_fundus
Tasks	grading, classification, quality
Samples	2,256
Classes	5 (No DR, Mild NPDR, Moderate NPDR, Severe NPDR, Proliferative DR)
Splits	train, val, test
Size	3.0 GB
Source-stated terms	CC BY-SA 4.0
Normalized terms	`cc-by-sa`
Descriptive screening label	Standard label without an explicit NC clause; not a permission finding
Terms scope	`dataset_files`
Access friction	`anonymous_direct`
Route backend	Zenodo
Availability	`available` (checked 2026-07-21)
Acquisition support	`standard_platform_supported`
Legacy sample-loader status	Standard loader included

Notes

Contains both regular fundus (2000 images in dual-field pairs) and ultra-widefield fundus (256 images). Labels CSV includes DR grade (0-4) and quality score.

Access preflight and acquisition

CLI
Python

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

# Explicit transfer, only when preflight reports supported behavior
eyehub download deepdrid --data-dir ./data

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

ds = REGISTRY.get_dataset('deepdrid')
print(preflight_dataset(ds, './data'))  # no transfer

Upstream page: zenodo.org/records

Source-term evidence: zenodo.org/records

Loader example

This entry includes a standard DatasetSample loader.

from pathlib import Path
from eyedatahub.datasets.registry import REGISTRY

data_dir = Path('~/.eyedatahub/data').expanduser()
ds = REGISTRY.get_dataset('deepdrid')
samples = ds.load(data_dir, split='test')
for s in samples[:5]:
    print(s.sample_id, s.label, s.image_path)

Citation

BibTeX
Plain text

@misc{deepdrid,
  title  = { DeepDRiD: Diabetic Retinopathy Grading and Image Quality Dataset },
  note   = { Liu R et al., 'DeepDRiD: Diabetic Retinopathy—Grading and Image Quality Estimation Challenge', Patterns 2022. Zenodo: https://zenodo.org/records/8248825 },
  year   = { 2022 },
  url    = { https://zenodo.org/records/8248825 },
}

Liu R et al., 'DeepDRiD: Diabetic Retinopathy—Grading and Image Quality Estimation Challenge', Patterns 2022. Zenodo: https://zenodo.org/records/8248825

Source-stated terms

Raw source string: CC BY-SA 4.0
Normalized category: cc-by-sa
Apparent scope: dataset_files
Descriptive screening label: Standard label without an explicit NC clause; not a permission finding

⚠️ 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.

mmrdr: MMRDR: Multi-Modal Retinal Diabetic Retinopathy Dataset (24,460 records, cc-by)
fundus_report_dataset: Fundus Report Dataset (422 records, cc-by)
airogs: AIROGS: AI for Robust Glaucoma Screening (113,893 records, cc-by-nc-nd)
eyecare_100k: Eyecare-100K: Multimodal Ophthalmology VQA Corpus (102,000 records, unknown)
justraigs: JustRAIGS: Just Referral AI Glaucoma Screening Dataset (101,442 records, cc-by-nc-nd)
eyepacs: EyePACS — Diabetic Retinopathy Detection (Kaggle 2015) (88,702 records, research-only)
multieye: MultiEYE: OCT-Enhanced Fundus Multi-Disease Benchmark (58,036 records, mit)
angioreport: AngioReport Fundus Angiography Report Dataset (55,361 records, unknown)

At a glance​

Notes​

Access preflight and acquisition​

Loader example​

Citation​

Source-stated terms​

Related datasets with shared modalities​