Post

New research · Ophthalmology
Translational vision science & technology · 17h
AI / informaticsTranslational vision science & technology · 2026

Artificial Intelligence-Driven Differentiation Between Uveal Melanoma and Nevus Based on Fundus Photographs: A Systematic Review and Meta-Analysis.

Theofilos Kanavos, Effrosyni Birbas, Jasmine H Francis … Theodoros P Zanos
Read paper
OphthalmologyAI / informatics

Artificial intelligence models distinguish uveal melanoma from nevus on fundus photos with high accuracy

Artificial Intelligence-Driven Differentiation Between Uveal Melanoma and Nevus Based on Fundus Photographs: A Systematic Review and Meta-Analysis.

Theofilos Kanavos … Theodoros P Zanos
Translational vision science & technology · 2026
Background

Distinguishing uveal melanoma (uveal melanoma) from uveal nevus (uveal nevus) is often challenging yet crucial for appropriate management.

Purpose

This study aimed to examine the ability of artificial intelligence (artificial intelligence) models to differentiate uveal melanoma from uveal nevus based on fundus photographs.

Methods

We systematically searched four databases through July 6, 2025, for studies developing machine learning models for distinguishing uveal melanoma from uveal nevus using fundus photographs as input.

n = 6208 participants
Results

Artificial intelligence correctly told melanoma from a harmless mole with excellent overall accuracy

Overall pooled
0.915AUC
Externally validated
0.873AUC
More results

Our review included seven articles with 6208 participants in total.

Six studies used deep learning and one applied conventional machine learning.

Only two studies conducted external validation.

More results

In subgroup analysis, externally validated models retained a promising discriminative ability with a subtotal pooled AUC of 0.873.

“
Conclusion · 1 of 3

Machine learning algorithms showcase consistently high performance in differentiating uveal melanoma from uveal nevus based on fundus photographs, supporting their potential role as adjunctive tools in clinical practice.

Conclusion · 2 of 3

Their objective, reproducible assessments may improve referrals, guide clinical decision-making, and boost diagnostic confidence across health care providers.

Conclusion · 3 of 3

However, existing evidence is highly heterogeneous and constrained by small dataset sizes and limited external validation.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
Cross-sectional
44.1%
had too much homocysteine in the blood, a far larger share than in dry AMD or healthy eyes
Cohort Study
92.3%
most laser-treated eyes had a favorable eye-structure outcome
Cohort Study
56%
lost three or more lines on the eye-chart vision test
Cohort Study
50%
older patients were more likely to lose three or more lines of vision
Study
81%
trials where mask air leaks were detected blowing toward the eyes
Cohort Study
13.5%
Central retinal artery peak blood-flow speed was 13.5% lower in diabetic patients.
Study
379 +/- 156 microm
lower thickness in the central retina, checked with an eye scan
Case Report
9 of 10 eyes
eyes had better measured vision after treatment