Is Retinal Age Ready for Clinical Use?
A simple photograph of the retina can now be analysed by artificial intelligence to estimate how old the retina appears.
Research suggests that people whose retinas appear older than expected may also have less favorable metabolic and cardiovascular profiles and higher long-term health risks.
This makes retinal age potentially interesting for preventive and longevity medicine.
But what can we actually do with this information today?
What has already been shown?
A large 2026 study used artificial intelligence to estimate age from retinal photographs of more than 71,000 UK Biobank participants.
The estimated retinal age closely matched chronological age.
Researchers then tested the model in an independent population from the Rotterdam Study, where it continued to estimate chronological age with good accuracy.
This is known as external validation.
It is important because it shows that the model’s ability to estimate age was not limited to the original UK Biobank population.
What exactly was validated?
The distinction is important.
External validation showed that the AI model could estimate chronological age from retinal photographs in another population.
It did not establish retinal age as a validated measure of the biological age of the whole body.
If a 60-year-old person receives a retinal-age estimate of 66, this does not mean that the heart, brain, kidneys, blood vessels and other organs are biologically 66 years old.
It means that the retinal features detected by that particular model appear older than expected for someone aged 60.
The important question is whether this difference carries useful information about health.
What can an older retinal age tell us?
In the 2026 study, a higher retinal age gap was associated with several established health risks, including higher blood pressure, poorer glucose control, greater body weight, smoking exposure and inflammatory markers.
It was also associated with several future health outcomes, including cardiovascular disease, stroke, dementia and all-cause mortality.
This does not mean that retinal age causes these diseases.
Instead, an older-appearing retina may reflect some of the vascular, metabolic and inflammatory processes that also influence health elsewhere in the body.
What should happen if retinal age is higher than expected?
An older retinal age should not be treated as a disease in itself.
Instead, it may provide another reason to look carefully at established and treatable health risks.
Depending on the individual, this may include assessment of:
blood pressure and hypertension,
glucose control and diabetes,
cholesterol and overall cardiovascular risk,
smoking,
body weight and metabolic health,
cardiovascular disease,
and eye health.
If abnormalities are identified, they should be managed according to established clinical evidence.
For example, hypertension should be appropriately treated, diabetes controlled, cardiovascular risk assessed and managed, and relevant retinal or ophthalmological disease evaluated.
The goal is therefore not to treat a retinal-age number. It is to identify and manage the underlying health risks that may be contributing to an unfavorable retinal profile.
What remains uncertain?
We still do not know whether retinal age itself should become a treatment target.
For example, if a person’s retinal age decreases after better blood-pressure control, weight loss, improved glucose control or another intervention, we cannot yet assume that this change independently means that biological aging has slowed.
We also do not yet know whether repeatedly measuring retinal age improves patient outcomes compared with using established clinical assessments alone.
In the 2026 study, retinal age provided some additional information beyond conventional risk factors, but the improvement in prediction was generally modest.
This is why retinal age should currently complement clinical assessment rather than replace it.
Where could retinal age become useful?
Retinal photography has several advantages.
It is rapid and non-invasive and provides direct images of vascular and neural tissue. Artificial intelligence can detect patterns within these images that may be difficult to recognize during conventional examination.
In the future, retinal age could therefore become another piece of information within a broader assessment of healthy aging.
Its value may not be in providing a single number that tells us “how old the body is.”
Its value may instead be in helping identify people whose retinal appearance suggests that cardiovascular, metabolic, inflammatory or eye-related health deserves closer attention.
That is a more realistic clinical interpretation of what the evidence currently supports.
