Earth logo

I can read the sign of early death in my eyes.

Believe in science.

By gaisndm HawkshawPublished 3 years ago 5 min read

A person's actual age does not necessarily faithfully reflect the degree of aging of the body.

But the eyes might be able to...

A person's actual age (chronological age) is calculated from the date of birth.

However, just because you are born for the same time doesn't mean you are as old as you are.

Some people are 40 years old, and their physical functions are comparable to those of ordinary people in their 20s, while others are 40 years old, and their organs function as if they were about 60 years old.

If you put aside the actual age, the degree of aging of the human body can also be observed in other ways.

For example, the overall decline in the level of DNA methylation in the genome is a sign of aging.

As a result, people often regard the degree of DNA methylation as an "aging clock".

Now, the eyes can also be used as a window to observe the aging process and even to assess the risk of death.

Recently, scientists have found that people who are older than their retina are more likely to die in recent years than others.

The team published the findings in the British Journal of Ophthalmology (British Journal of Ophthalmology).

But how can scientists tell how old a person's retina is?

AI can see the age of the retina.

When a person grows old, the running state of various organs in the body will change, from vitality to disrepair, and the eyes are no exception.

If you analyze the fundus images, you may find that the appearance of the retina varies with people of different ages.

With this in mind, scientists from the Guangdong Academy of Medical Sciences and the University of Melbourne have developed a deep learning AI to predict the age of the owner based on what the retina looks like in the fundus image.

To that end, the team collected 19,200 fundus images from the British Biological sample Bank (UK Biobank).

The pictures come from 11,052 people, ranging in age from 40 to 69, but have one thing in common: none of them had a previous medical history before the image was taken.

Scientists hope that AI will learn how the retina changes with age in healthy people.

Among the more than 10, 000 fundus images, most of the pictures are fed to AI as training data, and the actual age is marked for AI to explore the rules.

After the training is completed, the remaining pictures will be input to AI as verification data (equivalent to examination questions), without indicating the actual age, and let AI predict the age according to the picture.

The team found that the result of AI's prediction was very close to the actual age of the owner of the picture, with an error of less than 3.55 years.

This not only means that AI training is effective, but also means that the retina information contained in the image can reflect the actual age of human beings to a large extent.

In other words, the age of the retina of healthy people coincides with the actual age.

Of course, the study of these healthy people's data is just to make a "yardstick" to measure the situation of other people.

Scientists are more concerned about people who are not in good health, especially those whose retinas are older than they are.

What do you know about the risk of death?

After training, the qualified AI is also accepted. It is necessary to observe more people's fundus maps and give age prediction.

This time the image is still from the British Biological sample Bank, but instead of screening out healthy people, they are selected with or without a medical history.

The scientists collected fundus pictures of 35,913 people and knew how old they were at the time the pictures were taken and their health (including life and death) in the 11 years after the shooting.

The team gave these fundus maps to AI to predict, and then subtracted the age of the retina predicted by AI from the actual age to determine the age difference.

It was found that the greater the difference, the higher the probability that the owner of the picture would die within 11 years after the picture was taken.

Specifically, with each year of enlargement of the difference between the retinal age and the actual age, the probability of death of the image owner during the 11-year observation period increased by 2%.

In other words, the older the retina is, the higher the probability of death and the less likely it is to survive.

Among them, people whose retinas are more than 3 years older than their actual age are called "fast ager" by scientists.

Of the more than 30, 000 people surveyed, 1871 died during the observation period.

Of these, 321 (17.2%) could be attributed to cardiovascular disease, 1018 (54.4%) to cancer, and the remaining 532 (28.4%) to other causes.

And these situations can be discussed separately.

If you put aside cardiovascular disease and cancer, and only look at other parts, each year when the difference between the age of the retina and the actual age increases, the probability of death increases by 3%.

If you rank the difference from small to large, those in the bottom 25% are 67% more likely to die from "other causes" than those in the top 25%.

When looking at cases of cardiovascular disease alone, the scientists found that the relationship between the difference between retinal age and actual age and mortality was not significant.

But the team believes that this does not mean that the retina does not reflect the cardiovascular aging process, but with the development of medicine, some once fatal cardiovascular problems can be saved, reducing the mortality rate of related diseases.

As a result, scientists believe that the difference between the age of the retina and the actual age can be used as a biomarker of aging (biomarker) and can also be used to predict the risk of death.

Is there a scientific basis for this?

If it is only the result predicted by AI, without a scientific explanation, it may also be difficult to use retinal images to predict the degree of aging.

So, what health problems can the retina in the image show?

For example, kidney disease often goes hand in hand with eye disease.

After all, eyes and kidneys have many similarities in structure, development, and genetic pathways: the choroid and glomeruli in the eyes have structurally similar vascular networks, and the filtration barriers in the retina and glomeruli have similar developmental pathways.

Previous studies have shown that chronic nephropathy is associated with retinal maculopathy, and other studies have shown that abnormal retinal microvascular parameters can be used to predict chronic nephropathy.

The emboli in the retina, which can block blood vessels, have also been confirmed by scientists to be associated with many diseases such as chronic kidney disease.

Science

About the Creator

gaisndm Hawkshaw

Reader insights

Be the first to share your insights about this piece.

How does it work?

Add your insights

Comments

There are no comments for this story

Be the first to respond and start the conversation.

Sign in to comment

    Find us on social media

    Miscellaneous links

    • Explore
    • Contact
    • Privacy Policy
    • Terms of Use
    • Support

    © 2026 Creatd, Inc. All Rights Reserved.