Exploring AI's Potential in Papilledema Diagnosis to Support Dermatological Treatment Decisions in Rural…

AI / informaticsDiagnostics (Basel, Switzerland)· Oct 2025

Exploring AI's Potential in Papilledema Diagnosis to Support Dermatological Treatment Decisions in Rural Healthcare.

Jonathan Shapiro, Mor Atlas, Naomi Fridman … Idit Maharshak

Read paper

DermatologyAI / informaticsDiagnosis

GPT-4o achieved 85.9% accuracy in detecting papilledema from fundus images.

Exploring AI's Potential in Papilledema Diagnosis to Support Dermatological Treatment Decisions in Rural Healthcare.

Jonathan Shapiro … Idit Maharshak

Diagnostics (Basel, Switzerland) · Oct 2025

Background

Papilledema, an ophthalmic finding associated with increased intracranial pressure, is often induced by dermatological medications, including corticosteroids, isotretinoin, and tetracyclines.

Purpose

The primary objective of this study was to explore the diagnostic capability of ChatGPT-4o, a general large language model with multimodal input, in identifying papilledema from fundus photographs.

Methods

A dataset of 1094 fundus images (295 papilledema, 799 normal) was preprocessed and partitioned into a training set and a test set.

n = 1094 fundus images

Results

GPT-4o achieved moderate diagnostic accuracy detecting papilledema from fundus images

85.9%

GPT-4o

99.5%

ResNet

96%

Human

More results

For context, the ResNet model, fine-tuned for ophthalmic imaging, reached near-perfect accuracy (99.5%, Kappa = 0.99), while two human ophthalmologists achieved accuracies of 96.0% (Kappa ≈ 0.92).

“

Conclusion · 1 of 2

This study explored the capability of GPT-4o, a large language model with multimodal input, for detecting papilledema from fundus photographs.

Conclusion · 2 of 2

GPT-4o achieved moderate diagnostic accuracy and substantial agreement with the ground truth, but it underperformed compared to both a domain-specific ResNet model and human ophthalmologists.

Exploring AI's Potential in Papilledema Diagnosis to Support Dermatological Treatment Decisions in Rural Healthcare.

Exploring AI's Potential in Papilledema Diagnosis to Support Dermatological Treatment Decisions in Rural Healthcare.

Related papers