Authors
Thomas Scott Armstrong, Abdullahi Mohamed, Ojas Srivastava, Matthew T S Tennant
Published in
Canadian journal of ophthalmology. Journal canadien d'ophtalmologie. Aug 21, 2026. Epub Aug 21, 2026.
Abstract
Artificial intelligence (AI) is increasingly utilized for screening within ophthalmology, yet its application in rural communities remains poorly understood. This scoping review mapped modalities, feasibility, and barriers to implementation of AI in rural ophthalmology.
A scoping review (adhering to PRISMA-ScR guidelines).
A comprehensive literature search was conducted in August 2025. Eligible articles were original studies detailing clinical applications of AI in ophthalmology within a rural setting.
Eleven (78.6%) focused on diabetic retinopathy (DR), 2 (14.3%) focused on glaucoma, and 1 (7.1%) on multiple retinal lesions. AI modalities included cloud-based platforms and offline models. Sensitivities ranged from 81% to 100% with specificities of 84%-100% for DR screening, and 91% and 93%, respectively, in glaucoma detection. Multilesion screening demonstrated an area under the curve of 0.918. Patient acceptance ranged from 94% to 98.6%. Economic models supported the cost-effectiveness of AI in DR screening but not in glaucoma screening. Barriers included unreliable connectivity, lack of trained ophthalmic personnel, equipment fragility, and poor image quality.
AI is a valuable ophthalmological tool in the rural setting and could benefit geographically dispersed nations like Canada by amplifying disease screening and increasing accessibility of care. Use of this technology in rural locations can be complemented by longitudinal studies and technological innovation.
PMID:
42628931
Bibliographic data and abstract were imported from PubMed on 22 Aug 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 5
- Comments 0