Currently submitted to: JMIR AI
Date Submitted: Mar 12, 2026
Open Peer Review Period: Mar 20, 2026 - May 15, 2026
(currently open for review)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Evaluation of Retinal Findings in Dialysis Patients Using Artificial Intelligence: A Prospective Multicenter Study
ABSTRACT
Background:
Detecting eye complications early in dialysis patients is vital to avoid vision loss.
Objective:
This study aims to evaluate retinal findings in dialysis patients using artificial intelligence (AI) and to compare the diagnostic accuracy of AI with assessments made by two ophthalmologists. It also investigates the correlation between the severity of ocular conditions and factors such as dialysis vintage, medication usage, and systemic comorbidities.
Methods:
Prospective, multicenter study of 371 dialysis patients, retinal images were analyzed using the Eyecheckup AI system. Two independent ophthalmologists evaluated the same images. Primary outcomes included the detection of diabetic retinopathy, age-related macular degeneration, retinal vein occlusion, retinal anomalies, and glaucoma.
Results:
The agreement between AI, Doctor-1 and Doctor-2 was found in all measurements (p<0.05), with kappa values indicating substantial agreement for DR (73%), moderate agreement for ARMD (47.9%), moderate agreement for RVO (39.6%), substantial agreement for retinal anomalies (64.4%), and substantial agreement for glaucoma (74.5%). The agreement between the three measurements was as follows: for DR, there was a 65.5% (significant agreement), for ARMD, a 29.5% (moderate agreement), for RVO, a 46.7% (moderate agreement), for retinal anomalies, a 74.5% (significant agreement), and for glaucoma, an 80% (significant agreement) was observed. Significant differences were observed in dialysis vintage for DR (p=0.014) and Kt/V values for DR (p<0.001). Patients with DR had shorter dialysis vintage and lower Kt/V values compared to those without DR.
Conclusions:
AI demonstrates potential as a screening tool for ocular complications in dialysis patients, particularly for diabetic retinopathy and glaucoma.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.