JSIEC Fundus Photo Dataset

1,000 fundus images spanning 39 ophthalmic disease categories from the Joint Shantou International Eye Center. Used for multi-class fundus disease classification.

At a glance

Field	Value
Short name	jsiec
Full name	JSIEC Fundus Photo Dataset
First published	2019-10-08
Publication date precision	day
Publication date evidence	zenodo.org/api
Publication date source field	metadata.publication_date (earliest repository version)
Publication date reviewed	2026-09-11
Primary category	fundus
Resource role	current_dataset
Dataset family	jsiec
Contained modalities	fundus
Tasks	classification
Primary reported quantity	1,000 images
Classes	39 (0.0.Normal, 0.1.Tessellated fundus, 0.2.Large optic cup, 0.3.DR1, 1.0.DR2, 1.1.DR3, 10.0.Possible glaucoma, 10.1.Optic atrophy, 11.Severe hypertensive retinopathy, 12.Disc swelling and elevation, 13.Dragged Disc, 14.Congenital disc abnormality, 15.0.Retinitis pigmentosa, 15.1.Bietti crystalline dystrophy, 16.Peripheral retinal degeneration and break, 17.Myelinated nerve fiber, 18.Vitreous particles, 19.Fundus neoplasm, 2.0.BRVO, 2.1.CRVO, 20.Massive hard exudates, 21.Yellow-white spots-flecks, 22.Cotton-wool spots, 23.Vessel tortuosity, 24.Chorioretinal atrophy-coloboma, 25.Preretinal hemorrhage, 26.Fibrosis, 27.Laser Spots, 28.Silicon oil in eye, 29.0.Blur fundus without PDR, 29.1.Blur fundus with suspected PDR, 3.RAO, 4.Rhegmatogenous RD, 5.0.CSCR, 5.1.VKH disease, 6.Maculopathy, 7.ERM, 8.MH, 9.Pathological myopia)
Splits	all
Size	0.4 GB
Source-stated terms	Other open access (Zenodo; no standard license identifier)
Normalized terms	unknown
Descriptive screening label	Unknown or unclear; do not assume permission
Terms scope	unknown
Access friction	self_service_authenticated
Route backend	Zenodo
Availability	available (checked 2026-07-21)
Acquisition support	transfer_tested_partial
Legacy sample-loader status	Standard loader included

Reported quantities

Role	Count	Unit	Scope	Basis	Evidence
Primary	1,000	images	Primary quantity reported in the reviewed catalog source	legacy_catalog_field	zenodo.org/record

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

Zenodo labels the deposit as other open access but does not name a standard reuse license. Verify terms before redistribution.

Documented relationships

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

• rao_fundus is derived from this record: The RAO source reports use of public web images plus RFMiD and JSIEC images. (evidence)
• smdg is derived from this record: The official SMDG source table lists this catalog record among the 19 standardized source domains. (evidence)
• x_pcr is derived from this record: Source labels in the version-pinned public X-PCR deposit identify this catalog record as upstream material. (evidence)

Access information and download

CLI

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

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

Python

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

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

Upstream page: zenodo.org/record

Source-term evidence: zenodo.org/record

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('jsiec')
samples = ds.load(data_dir, split='all')
for s in samples[:5]:
    print(s.sample_id, s.label, s.image_path)

Citation

BibTeX

@misc{jsiec,
  title  = { JSIEC Fundus Photo Dataset },
  note   = { Cen et al., 'Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks', Nature Communications 2021. doi:10.1038/s41467-021-25138-w. Data: doi:10.5281/zenodo.3477553 },
  year   = { 2021 },
  url    = { https://zenodo.org/record/3477553 },
}

Plain text

Cen et al., 'Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks', Nature Communications 2021. doi:10.1038/s41467-021-25138-w. Data: doi:10.5281/zenodo.3477553

Source-stated terms

• Raw source string: Other open access (Zenodo; no standard license identifier)
• Normalized category: unknown
• Apparent scope: unknown
• Descriptive screening label: Unknown or unclear; do not assume permission

⚠️ 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)