Clinical Characteristics of Healthy Eyes for Glaucoma AI Validation

370 eyes from 193 patients, including 122 healthy eyes, were used to evaluate two AI systems for glaucoma detection. Gilvan Vilarinho da Silva Filho published this retrospective cross-sectional study on figshare in 2026. Both Laguna ONhE and VUNO Med-Fundus AI systems demonstrated high diagnostic accuracy using color fundus photographs.

Use Cases

• Benchmarking AI diagnostic performance based on reported AUC metrics for glaucoma detection
• Training or validating models for glaucoma classification using clinical eye characteristics
• Studying the combined effect of multiple AI systems on diagnostic accuracy as described in the results

Strengths

• Dataset includes a specific count of 370 eyes from 193 patients, with 122 healthy eyes and 248 glaucomatous eyes.
• Diagnostic accuracy metrics are provided, including AUC values of 0.879 and 0.857 for the two AI systems.
• Data is shared under a permissive CC-BY-4.0 license, facilitating reuse.

Limitations

• Row count and column-level documentation are unknown, which may limit suitability assessment.
• The dataset is very small at 5.5 KB, indicating limited scope.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Retrospective cross-sectional study using swept-source optical coherence tomography and visual field testing.

Freshness

Last updated 2026-05-28 17:34:33; freshness should be verified.