Humans are highly dependent on vision animals, in daily life we receive more than 80% of the information from the eyes, worthy of the title of human "most important body surface organs". People often say that the eyes are the windows of the soul, but they do not know that the eyes are the windows of the body, and can even predict the disease, aging and death of the human body.
As we all know, the anti-aging community has always been "whole life", a study that combines ophthalmology and artificial intelligence, led by He Mingguang, professor of ophthalmology at the University of Melbourne in Australia and second-level professor of the Ophthalmology Center of Sun Yat-sen University, pointed out:
"The retinal age gap can serve as a new marker of aging that more effectively predicts the health and risk of death in the body, as evidenced by a 2 percent increase in the risk of all-cause death and a 3 percent increase in the risk of death from specific causes other than cardiovascular disease and cancer for every 1 year of increase in the retinal age gap."
The study was published in the British Journal of Ophthalmology in 2022.[1]
Note: The retinal age gap can be used as a biomarker to predict mortality
What is the "retinal age gap"
Although it may seem to be tortuous, the concept of the retinal age gap is not unfamiliar. The Time School has been in "Chinese and Foreign Scholars' Heavy Discoveries! A new method for detecting the age of human organs was released, and the era of multi-dimensional and accurate anti-aging is coming (click on the blue word jump) summarizes and reports on the multiple "clocks" in the body that regulate the aging process - the so-called biological age. The retina also has its own age, and the difference between the age of the retina and the actual age of the individual is called the age difference of the retina.
Scientists are not on a whim by setting their sights on the retina, which has similar embryonic origins, physiological features and anatomy to vital organs such as the heart, brain and kidneys. As the only organ that can observe microvascular and neural tissue in the body, the retina faithfully reflects the state of blood circulation and has the same neurodegenerative pathological changes as the brain. In addition, fundus photography and artificial intelligence analysis can capture the retinal age gap, a fast, economical, non-invasive feature that no other "clock" can match.
Heaven and time, location, people and people
Professor He Mingguang, the corresponding author of this paper, is currently the leader of the research group of the State Key Laboratory of Ophthalmology of Sun Yat-sen University, and has become well-known in the field of international artificial intelligence ophthalmology, and the AI ophthalmologist system created by his team has also reached the world's leading level. The profound ophthalmic strength coupled with advanced artificial intelligence algorithms can be described as taking advantage of the time and place, and the UK biobank has handed over the last piece of the "people" puzzle.
The UK Biobank collected nearly half a million volunteers aged 40-69 between 2006 and 2010, including a comprehensive health questionnaire, detailed body measurements and collection of biological samples, and conducted years of follow-up surveys, and the database is currently available for researchers worldwide to apply for use[2].
Deep learning algorithms "catch the eye"
The team first screened 19,200 fundus images from 11,052 healthy participants from the database and fed them to the AI, and the results showed that there was a strong correlation between retinal age and actual age, with overall accuracy within 3.5 years, better than most existing biomarkers of aging: for example, the average standard error of DNA methylation clocks was 3.3-5.2 years, the blood spectrum clock was 5.5-5.9 years old, and the transcriptome aging clock was 6.2-7.8 years [3-8].
After a large amount of data correction, the deep learning algorithm analyzed 35917 fundus pictures from 35913 random volunteers and calculated the retinal age gap (retinal age - actual age), a positive number representing the retina "more aging" and a negative number indicating that the retina is "younger".
Note: Deep learning algorithms assess retinal age gaps
In order to more accurately assess the impact of the retinal age gap, the researchers optimized the algorithm model for the covariate variables of the population's age, race, sex, education level, smoking status, diabetes history, heart disease history, etc., and concluded that "for every 1 year increase in the retinal age gap, the risk of all-cause death increases by 2%, and the risk of death from specific causes other than cardiovascular disease and cancer increases by 3%."
In an average of 11 years of follow-up visits, a total of 1871 volunteers died of all-cause deaths, of which 321 (17.2%) died of cardiovascular disease, 1018 (54.4%) died of cancer, and 532 (28.4%) died of other specific causes.
Among them, the correlation between specific causes of death and the age gap of the retina reached 49%-67%, while the correlation between cardiovascular disease and cancer was not significant, which may be due to the fact that modern medicine has a more developed treatment for cardiovascular disease and cancer, which greatly reduces mortality.
Figure note: Association between the retinal age gap and the risk of death
Beyond the data, what kind of future
The main body of the paper is roughly like this, and the authors also review the clinical significance of the study at the end: first, the retinal age gap can be used as a novel and accurate biomarker of aging; second, the rapid, economical, and noninvasive nature of fundus imaging makes it a convenient screening tool for identifying individuals with an increased risk of death.
With the development of smartphones and the integration of deep learning algorithms, in the future, perhaps each of us will be able to detect our own retinal age anytime, anywhere, detect a variety of eye disease conditions, and make predictions about the risk of other diseases.
Despite the study's large sample sizes, long-term follow-up, standardized fundus image acquisition protocols, and adjustments to multiple confounding factors in statistical models, the authors acknowledge that the studies had limitations, such as that these retinal images were captured at specific moments and that participants were not representative of the entire population because people with very poor health were less likely to participate in the study.
Time Pie reviews
Since the UK Biobank made the data public, a variety of studies have sprung up. It has to be acknowledged that this is a great undertaking, and the data it provides, if used properly, can greatly improve human health and health care. At the same time, I also expect that China can also have its own large-scale biological sample bank, so as to improve the national health level more accurately. If you have unlimited interest and curiosity in anti-aging research, you can enter the exchange club skirt, pai assistant v: timepie10. Returning to this article, despite the limitations of the study, the authors first proposed the concept of "retinal age gap" and integrated deep learning algorithms to evaluate its association with mortality, providing a reliable reference for follow-up studies and a future that people can imagine.
—— TIMEPIE ——
There are more anti-aging science content in the homepage, welcome to browse and pay attention to ~
References (too long, if necessary, you can ask the assistant)