LAG: Large-scale Attention-based Glaucoma Database

11,760 fundus images with glaucoma classification labels and ophthalmologist-derived attention maps. Largest public glaucoma dataset with attention ground truth.

At a glance

Field	Value
Short name	lag
Full name	LAG: Large-scale Attention-based Glaucoma Database
First published	Unknown
Publication date precision	Unknown
Publication date evidence	Unknown
Publication date source field	Unknown
Publication date reviewed	Unknown
Primary category	fundus
Resource role	current_dataset
Dataset family	lag
Contained modalities	fundus
Tasks	classification, segmentation
Primary reported quantity	11,760 images
Classes	2 (normal, glaucoma)
Splits	train, test
Size	3.0 GB
Source-stated terms	Research only (no redistribution; password by email)
Normalized terms	research-only
Descriptive screening label	Research or challenge restriction recorded; check source
Terms scope	dataset_files
Access friction	author_contact
Route backend	Manual (upstream-gated)
Availability	available (checked 2026-07-21)
Acquisition support	manual_access_blocked
Legacy sample-loader status	Metadata and access only

Reported quantities

Role	Count	Unit	Scope	Basis	Evidence
Primary	11,760	images	Primary quantity reported in the reviewed catalog source	legacy_catalog_field	github.com/smilell

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

Dropbox link gated by password — email [email protected] to request access. Cannot be auto-downloaded.

Documented relationships

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

• glaucoma_expert_cot_refined is derived from this record: The official dataset card identifies LAG and PAPILA as the source fundus resources paired with the refined glaucoma reasoning records. (evidence)
• mm_retinal_reason is derived from this record: The version-pinned official dataset card lists this record among the CFP or OCT sources used to construct MM-Retinal-Reason. (evidence)

Access information and download

CLI

# This route requires upstream human action; no transfer starts.
eyehub download lag --data-dir ./data --dry-run --json
# Follow the official instructions shown by preflight.

Python

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

ds = REGISTRY.get_dataset('lag')
print(preflight_dataset(ds, './data'))  # returns manual_access_blocked

Upstream page: github.com/smilell

Source-term evidence: github.com/smilell

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{lag,
  title  = { LAG: Large-scale Attention-based Glaucoma Database },
  note   = { Li et al., 'Attention Based Glaucoma Detection: A Large-scale Database and CNN Model', CVPR 2019 },
  year   = { 2019 },
  url    = { https://github.com/smilell/AG-CNN },
}

Plain text

Li et al., 'Attention Based Glaucoma Detection: A Large-scale Database and CNN Model', CVPR 2019.

Source-stated terms

• Raw source string: Research only (no redistribution; password by email)
• Normalized category: research-only
• Apparent scope: dataset_files
• Descriptive screening label: Research or challenge restriction recorded; check source

⚠️ 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)