ScholarDeck

AI / informaticsActa ophthalmologica · 2026

Deep learning in glaucoma referral: Performance assessment using a real-world setting.

Afonso Lima-Cabrita, Rafael Correia Barão, Diogo Bernardo Matos … Luís Abegão Pinto

Read paper

OphthalmologyAI / informatics

Deep learning model could prevent over a third of unnecessary glaucoma referrals.

Deep learning in glaucoma referral: Performance assessment using a real-world setting.

Afonso Lima-Cabrita et al. · Acta ophthalmologica · 2026

Purpose

Evaluate a deep learning model's performance as a pre-referral filter for referable glaucoma using colour fundus photographs.

Methods

Referred for glaucoma evaluation in 2021 were included.

n = 96 patients

Results

the model could prevent unnecessary glaucoma referrals

sensitivity 1 (95% CI 0.82 to 1)

CI excludes the null - significant

More results

Of these referred patients, 37 were accepted by the glaucoma department (38.5%); 19 were found to have glaucoma (19.8%).

The model presented high sensitivity (1.00; 95% CI 0.82-1.00), and reasonable specificity (0.65; 95% CI 0.53-0.75) and positive predictive value (0.41; 95% CI 0.27-0.57).

“

Conclusion · 1 of 2

Glaucoma rate in ophthalmologist-referred patients was low. A deep learning-led system would have accurately referred all patients with glaucoma and reduced unnecessary observations.

Conclusion · 2 of 2

Current model performance can act as a filter between referring and receiving specialised healthcare.

Post

Deep learning in glaucoma referral: Performance assessment using a real-world setting.

Deep learning model could prevent over a third of unnecessary glaucoma referrals.

Related papers