Post

New research · Ophthalmology
BMC ophthalmology · 13h
AI / informaticsBMC ophthalmology · 2026

AI-assisted diagnosis of neuro-ophthalmic disorders: a systematic review from optic neuritis to papilledema.

Shih-Shuan Fang, Sheng-Han Chen
Read paper
OphthalmologyAI / informatics

Artificial intelligence models detect papilledema on fundus photos with very high accuracy

AI-assisted diagnosis of neuro-ophthalmic disorders: a systematic review from optic neuritis to papilledema.

Shih-Shuan Fang, Sheng-Han Chen
BMC ophthalmology · 2026
Background

Neuro-ophthalmic disorders like optic neuritis and papilledema are diagnostically challenging, with non-specific symptoms and reliance on expert interpretation of optic nerve head imaging (Liu TYA et al.

Purpose

This systematic review evaluates the current state of artificial intelligence in diagnosing key neuro-ophthalmic disorders, focusing on optic neuritis and papilledema.

Methods

We extracted data on study design, artificial intelligence model architecture, and diagnostic accuracy, performing a qualitative synthesis of the findings.

n = 32 eligible studies
AUC > 0.98
Results
AUC > 0.98
detected swollen optic nerve on photos with near-perfect accuracy
n = 32 eligible studies
More results

Most (66%) focused on papilledema, while 34% addressed optic neuritis.

OCT was the most common imaging modality (75%).

In optic neuritis, artificial intelligence models detected chronic RNFL thinning effectively (AUC > 0.95) but struggled to differentiate the pattern of atrophy from other conditions like NAION or glaucoma (AUCs 0.85–0.92).

More results

A major gap was the lack of models for diagnosing acute optic neuritis.

“
Conclusion · 1 of 3

Artificial intelligence shows significant potential as a diagnostic aid in neuro-ophthalmology, particularly in detecting optic disc swelling and chronic optic atrophy. However, the field is in its early stages.

Conclusion · 2 of 3

Current models excel at binary classification but struggle with the core clinical problem of differential diagnosis.

Conclusion · 3 of 3

Future progress depends on curating large, multi-center datasets, developing sophisticated multimodal models, and focusing on solving the core clinical problem of differential diagnosis.

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