Four Clinical AI Diagnostic Tasks and How To Evaluate Them
Triage, detection, characterization, and autonomous diagnosis make different clinical claims. Here is the evidence question that belongs to each one.
Medical Specialty Hub
AI in Ophthalmology
AI for retinal image analysis, diabetic retinopathy screening, and eye disease detection.
Overview
Ophthalmology has produced some of the earliest FDA-cleared AI medical devices. AI systems can analyze fundus photographs to detect diabetic retinopathy, glaucoma, and age-related macular degeneration with high accuracy.
Ophthalmology holds a position no other specialty does: it is where autonomous AI became legal in American medicine.
The first autonomous diagnostic device authorized in the United States screens retinal photographs for diabetic retinopathy and returns a result with no clinician viewing the image. Every other device on the FDA list produces something a clinician interprets. This category produces a determination.
Why here first
Retinal photography is cheap, standardized, and repeatable. Diabetic retinopathy is common, has a clear grading standard, and has a screening question with a binary answer. And the clinical gap is enormous: a large share of diabetic patients never reach an eye clinic at all, so the comparator for autonomous screening is frequently no screening rather than specialist screening.
That last point is what made the risk calculation work. A device does not have to match an ophthalmologist to improve outcomes in a population that was not seeing one.
Clinical Use Cases
- Diabetic retinopathy screening
- Glaucoma detection
- Age-related macular degeneration analysis
- Retinal vessel analysis
- Autonomous AI diagnostic systems
Where The Evidence Actually Stands
This is the best-evidenced category in medical AI, with prospective validation at primary care sites using the intended non-specialist staff. The caveats are operational rather than technical: ungradable image rates are real, screening is indicated for one disease and does not assess other retinal pathology, and the benefit evaporates entirely if positives do not actually reach an ophthalmologist. Referral leakage is the recurring failure in real programs.
FDA-Authorized Devices In This Specialty
3 records currently tracked.
Related Research
FDA-Cleared AI Medical Devices Explained
What FDA clearance means for AI medical devices, how the regulatory pathways work, and what clinicians need to understand before deploying FDA-cleared AI tools.
How To Evaluate A Clinical AI Tool
Twelve questions that separate a clinical AI tool worth deploying from one that will look impressive in a demonstration and disappoint in production.