Glaucoma Detection: Retinal Fundus Images for Stage Classification

Glaucoma Detection is a dataset of retinal fundus images for glaucoma stage classification, created by moondream and last updated on April 7, 2026. It contains 2,847 training, 1,259 validation, and 1,272 test samples. Each sample includes an image and a label for one of three glaucoma stages: normal, early, or advanced.

Use Cases

• Train image classification models to detect glaucoma based on retinal fundus images.
• Develop multi-stage classifiers to distinguish between normal, early, and advanced glaucoma findings.
• Benchmark model performance on a medical imaging task with defined train/validation/test splits.
• Explore feature extraction techniques for diagnostic patterns in retinal imagery.

Strengths

• Dataset provides explicit splits for training (2,847 samples), validation (1,259), and testing (1,272).
• Class labels are clearly defined with descriptions for three distinct glaucoma stages.
• The dataset is focused on a specific, clinically relevant medical imaging task.

Limitations

• Row count for the full dataset is unknown, which may limit suitability assessment.
• Column-level documentation beyond 'image' and 'class' is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

huggingface user moondream

Collection Method

Collection method is not specified in the provided description.

Time Range

Temporal coverage is not specified.

Freshness

Last updated 2026-04-07 17:33:58; freshness should be verified.

Geography

Spatial coverage is not specified.