Pulse.

a daily field guide to health research that matters

◆ Console

‹ Mon · 3 Aug 2026
Promising but preliminary

Classification of Inherited Retinal Diseases Using Artificial Intelligence Models for Fundus Autofluorescence and Ultrawide Retinal Images.

AI trained on fundus images accurately identifies five inherited retinal diseases, potentially speeding diagnosis in eye care clinics.

This study adapted the RETFound foundation model and CNN architectures for classifying five inherited retinal diseases (Best disease, rod-cone dystrophy, Stargardt disease, choroideremia, and normal) from fundus autofluorescence and pseudocolour UWF images. ResNet18 achieved the best overall performance (F1 0.839), outperforming classical ML approaches, establishing a deployable framework for IRD triage in clinical eye care settings.

What the study was

Study design
retrospective_validation
Category
ai_ml_diagnostics_imaging
Maturity
Validated
Journal
J Ophthalmol

Why it surfaced

Rare hereditary diseases cause significant working-age blindness; AI-aided IRD classification using foundation models has direct clinical triage value; multimodal imaging approach adds robustness to the framework.

A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.