Post

New research · Ophthalmology
Eye (London, England) · 4d
AI / informaticsEye (London, England) · 2026

Clinically aligned artificial intelligence for glaucoma diagnosis: enhancing retinal nerve fibre layer interpretation from fundus images.

Shang-Lin Chung, Jehn-Yu Huang, Yi-Ching Shao … Chia-En Lien
Read paper
OphthalmologyAI / informatics

Artificial intelligence system correctly flags most fundus photos needing glaucoma specialist referral

Clinically aligned artificial intelligence for glaucoma diagnosis: enhancing retinal nerve fibre layer interpretation from fundus images.

Shang-Lin Chung … Chia-En Lien
Eye (London, England) · 2026
Background

This study aimed to develop an interpretable artificial intelligence (artificial intelligence) screening system that replicates a specialist's evaluation of fundus photographs.

Purpose

This study aimed to develop an interpretable artificial intelligence (artificial intelligence) screening system that replicates a specialist's evaluation of fundus photographs.

Methods

A total of 773 fundus images from the independent test cohort were annotated by three fellowship-trained glaucoma specialists, followed by repeated consensus meetings to improve annotation consistency.

n = 773 fundus images
Results

correctly identifies about 9 in 10 patients who truly needed specialist referral

Sensitivity 0.90 (95% CI 0.87 to 0.94)
null 0
0.87
0.94
CI excludes the null - significant
More results

The final independent test cohort comprised 268 referral and 481 non-referral cases, providing a basis for evaluation.

The system captured complementary aspects of glaucomatous optic neuropathy that are often missed by single-feature approaches.

“
Conclusion · 1 of 2

Consensus-based annotation, combined with lesion-level modelling, enhances alignment with clinical reasoning.

Conclusion · 2 of 2

By providing an explicit referral rationale, the system fosters trust in artificial intelligence-assisted glaucoma screening and facilitates adoption in clinical settings.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving