Truace
Subscribe

Review of 17 studies assesses AI screening for diabetic eye disease using wide retinal images

Study title: Ultra-widefield color fundus images and artificial intelligence for diagnosis of diabetic retinopathy: A systematic review and meta-analysis

A systematic review and meta-analysis to February 9, 2025, evaluated artificial intelligence for diabetic retinopathy assessment using ultra-widefield color fundus images, which capture a larger retinal area without pupil dilation. Of 527 records, 17 studies were reviewed and four were meta-analyzed, all using Optos software, to estimate sensitivity and specificity for AI-driven screening.

Read the study abstract

Ultra-widefield (UWF) fundus cameras capture a larger retinal area without pupil dilation. We summarized evidence and diagnostic performance of artificial intelligence (AI)-driven diabetic retinopathy (DR) assessments using UWF images (UWFIs). We searched PubMed, Scopus, the Cochrane Library, and Web of Science to February 9, 2025, for studies evaluating DR using UWFIs and AI analyses. We followed the PRISMA guidelines and assessed study quality using the Joanna Briggs Institute Critical Appraisal Checklist for…

By the Truvace desk1 min read1 sourceEvidence: Moderate · 70/100 — Systematic review or meta-analysis · 1 independent sourcePeer-reviewedTRV-2026-0564· cite & verify
Ultra-widefield color fundus images and artificial intelligence for diagnosis of diabetic retinopathy: A systematic review and meta-analysis
Hospital Universitari Doctor Peset, València 01 by 19Tarrestnom65. CC BY-SA 4.0 · https://creativecommons.org/licenses/by-sa/4.0
Trace impact readingContested
P 57The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.

Both sides are scored from claims and sources, not community votes.

G 57The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.

In brief

The pooled performance suggests AI can detect diabetic retinopathy from ultra-widefield images with acceptable sensitivity but lower specificity, indicating potential for screening but risk of false positives. Uncertainty remains because the evidence base is small and lacks rigorous external validation across diverse datasets and ultra-widefield platforms, limiting generalizability to broader clinical use.

Main points

  1. Systematic review searched PubMed, Scopus, Cochrane Library, and Web of Science to February 9, 2025, identifying 527 records with 17 included in review and four in meta-analysis.
  2. All included studies used Optos software for ultra-widefield imaging.
  3. Included studies covered DR discrimination from multiple disorders, stage classification, screening, referable pathology detection, progression prediction, and clinical feature detection.

The gain

AI-driven diabetic retinopathy screening using ultra-widefield fundus images achieved a summary sensitivity of 85.0% and AUC of 0.870 in meta-analysis.

The problem

AI-driven diabetic retinopathy screening using ultra-widefield fundus images showed limited specificity of 72.5% in meta-analysis.

The rundown

The review included 17 studies in the systematic review and four studies in the meta-analysis, all using Optos software, with study quality assessed using the Joanna Briggs Institute Critical Appraisal Checklist for diagnostic accuracy.

Study purposes varied: two evaluated DR discrimination from multiple retinal disorders, six addressed DR stage classification based on guideline scales, six evaluated DR screening, and one each focused on referable pathology detection, progression prediction, and clinical feature detection.

What this doesn’t fix

Findings are limited by small evidence base and lack of diverse external validation across platforms.

Sources

  1. Peer-reviewedRetina2026-07-24

The debate