Can AI detect certain diseases by looking at images of eyes ?
Cast your vote — then read what our editor and the AI models found.
AI systems are increasingly able to identify certain diseases by analyzing images of the retina. These tools examine retinal scans to detect conditions like diabetic retinopathy, glaucoma, and age-related macular degeneration, as well as broader health risks such as cardiovascular disease. How exactly are these models trained and what evidence supports their effectiveness?
Background
AI systems can analyze retinal images to detect diseases, particularly using retinal scans such as fundus photographs and optical coherence tomography (OCT). These systems have demonstrated high accuracy in identifying conditions including diabetic retinopathy, glaucoma, and age-related macular degeneration. Some models also predict systemic diseases like hypertension and cardiovascular risk from retinal images.
Deep learning models have shown strong performance for diseases such as diabetic retinopathy, age-related macular degeneration, glaucoma, and neurodegenerative conditions including Alzheimer’s disease, often matching or exceeding expert clinicians on specific diagnostic tasks. These models rely on large labeled datasets of fundus photographs, OCT scans, and sometimes multi-modal imaging to identify subtle vascular, structural, and texture changes linked to disease.
Regulatory-cleared tools based on these models are already in clinical use today. However, widespread adoption depends on validation across diverse populations and seamless integration into existing ophthalmic workflows.
— Enriched May 13, 2026 · Source: Nature Medicine — Enriched May 13, 2026 · Source: National Eye Institute
Suggest a tag
A missing concept on this topic? Suggest it and admin reviews.
Status last checked on June 23, 2026.
Gallery
Can AI detect certain diseases by looking at images of eyes?
The jury found a clear answer in the affirmative.
After weighing the evidence, the jury found that AI has crossed the threshold from promising prototype to clinically useful diagnostician in ophthalmology. The lone juror concluded that peer-reviewed trials now exceed the “toddler scribbling” phase and deliver real, reproducible diagnostic skill. Ruling: “The ophthalmoscope’s pupil now blinks back a diagnosis—verdict for the affirmative, unanimously.”
But the data is real.
The Case File
Across 9 sessions, 27 jurors have heard this case. Combined tally: 25 YES · 2 ALMOST · 0 NO · 0 IN RESEARCH.
Note: cumulative includes older juror opinions. The current session tally above is the live verdict.
By a vote of 1 — 0 — 0, the panel returns a verdict of YES, with verdict confidence of 98%. The court so orders.
"Specialized AI systems detect diabetic retinopathy, AMD, and glaucoma from retinal images with clinically validated accuracy."
What the audience thinks
No 0% · Yes 74% · Maybe 26% 23 votesDiscussion
no comments⚖ 9 jury checks · most recent 4 days ago
Each row is a separate jury check. Jurors are AI models (identities kept neutral on purpose). Status reflects the cumulative tally across all checks — how the jury works.
More in health
Can AI identify tuberculosis from cough audio recordings with better accuracy than human clinicians ?
Can AI predict the outcome of a clinical drug trial based on molecular structure alone ?
Can AI detect microplastic particles in seawater from drone-captured hyperspectral imagery ?