Affinity Health, a leading provider of high-quality health cover, is highlighting how artificial intelligence (AI) could help detect serious eye conditions earlier this World Sight Day (8 October).
Many serious eye conditions develop without pain, blurred vision, or clear warning signs. AI can help eyecare professionals examine images, detect small changes, and identify people who need further tests. It is important to note, however, that AI is not expected to replace regular eye examinations, at least not in the immediate future.
“The World Health Organization (WHO) estimates that at least 2.2 billion people worldwide have a near or distance vision impairment, and that at least one billion of these cases could have been prevented or are yet to be addressed. In our view, those figures show just how important early detection is” says Murray Hewlett, CEO of Affinity Health.
“It is easy to assume that our eyesight is fine when we are not noticing any problems. We find it encouraging that AI could help eyecare professionals identify possible concerns sooner, giving more people the opportunity to receive care before their vision is seriously affected.”
How Does AI Examine the Eye?
During an eye examination, a special camera can take a detailed photograph of the retina. The retina is the light-sensitive layer at the back of the eye that sends visual information to the brain.
Another test, called optical coherence tomography, or OCT, creates detailed images of the retina’s different layers. The scan is quick, painless, and can reveal swelling, fluid, or changes that are difficult to see during a basic examination.
AI systems are trained using large collections of eye images. The software learns to recognise disease-related patterns, including damaged blood vessels, unusual swelling, and changes to the optic nerve.
When it analyses a new image, the system can flag areas that look unusual. Depending on the AI tool, it may also help measure how serious a problem is or recommend referring the patient to an ophthalmologist for further tests.
The strongest evidence for AI in ophthalmology currently relates to analysing eye images.
Finding Diabetic Eye Disease Earlier
Diabetic retinopathy develops when high blood sugar damages the tiny blood vessels in the retina. It often causes no noticeable symptoms at first, but as it progresses, the vessels may leak blood or fluid, leading to blurred vision, dark patches, floaters, and, in severe cases, vision loss.
AI-assisted screening can analyse retinal photographs for small bleeds, swelling, and damaged blood vessels. An unusual result does not confirm the condition, but it may indicate that an eyecare professional should examine the patient further.
Which Other Conditions Can AI Help Identify?
Researchers are also developing and testing AI tools for several other eye conditions, including:
- Glaucoma, which damages the optic nerve and can gradually reduce vision.
- Age-related macular degeneration, which affects the central part of a person’s sight.
- Diabetic macular oedema, which causes fluid to build up and swell the retina.
- Cataracts, which develop when the eye’s natural lens becomes cloudy.
- Retinopathy of prematurity, which may affect premature babies.
- Keratoconus, which causes the clear front surface of the eye to become thinner and change shape.
Some systems can compare a patient’s latest scan with previous images. This allows eyecare professionals to monitor changes and decide whether treatment or more frequent check-ups may be required.
However, these technologies are not available at every optometrist, clinic, or hospital. Some tools have been approved for specific uses, while others are still being studied.
Could AI Make Eye Care More Accessible?
AI could help people who live far from an eye specialist. A photograph of the back of the eye can be taken at a clinic, mobile health unit, or screening centre and checked quickly using AI. People with possible signs of disease can then be referred for further care.
However, AI is not always correct. Blurred photographs and poorly trained systems can produce unreliable results. These tools must be properly tested and used by trained eyecare professionals.
Conclusion
AI can help eyecare professionals, but it cannot replace a full eye examination. Apps and online tests are not suitable for a proper diagnose of an eye condition. It is generally recommended to seek urgent medical help if you experience sudden vision loss, severe eye pain, flashes of light, new floaters, double vision, or a dark curtain across your sight. Affinity Health helps members care for their sight with one eye test and one set of standard frames and lenses per member per 24 months, available only through a Spec-Savers outlet and subject to plan terms, conditions, and waiting periods.
