New study finds ‘simple selfie' can help predict patients' cancer survival
Because humans age at 'different rates' their physical appearance may help give insights into their so-called 'biological age' – how old a person is physiologically, academics said.
The new FaceAge AI tool can estimate a person's biological age, as opposed to their actual age, by scanning an image of their face, a new study found.
A person's biological age, which is a predictor of their overall health and can be a predictor of life expectancy, is based on many factors including lifestyle and genetics, researchers from Mass General Brigham in the US said.
But they wanted to examine whether or not biological age could be examined based on how a person looks – similar to what doctors call an 'eyeball test' whereby certain judgments are made based on how a person looks, such as whether or not someone could undergo intensive cancer treatment based on how frail they appear to be.
Researchers said they wanted to see whether they could 'go beyond' the 'subjective and manual' eyeball test by creating a 'deep learning' artificial intelligence (AI) tool which could assess 'simple selfies'.
The new algorithm was trained using 59,000 photos.
'Our study now has shown for the first time that we can really use AI to turn a selfie into a real biomarker source of ageing,' said Dr Hugo Aerts, corresponding author of the paper.
He said the tool is low cost, can be used repeatedly over time and could be used to track an individual's biological age over 'months, years and decades'.
'The impact can be very large, because we now have a way to actually very easily monitor a patient's health status continuously and this could help us to better predict the risk of death or complications after, say, for example, a major surgery or other treatments,' he added.
Explaining the tool, academics showed how it assessed the biological age of actors Paul Rudd and Wilford Brimley based on photographs of the men when they were both 50 years old.
Rudd's biological age was calculated to be 42.6, while Brimley, who died in 2020, was assessed to have a biological age of 69.
The new study, published in the journal Lancet Digital Health, saw the tool used on thousands of cancer patients.
FaceAge was used on 6,200 patients with cancer using images taken at the start of their treatment.
The academics found that the biological age of patients with cancer was, on average, five years older than chronological age.
They also found that older FaceAge readings were associated with worse survival outcomes among patients with cancer, especially in people who had a FaceAge older than 85 years old.
The authors concluded: 'Our results suggest that a deep learning model can estimate biological age from face photographs and thereby enhance survival prediction in patients with cancer.'
Dr Ray Mak, co-senior author on the paper, added: 'We have demonstrated that AI can turn a simple face photo into an objective measure of biological age, that clinicians can use to personalised care for patients, like having another vital sign data point.'
He said that it is 'another piece of the puzzle like vital signs, lab results or medical imaging'.
But he added: 'We want to be clear that we view AI tools like FaceAge as assistance provide decision support and not replacements for clinician judgment.'
More studies assessing FaceAge are under way, including whether it could be used for other conditions or diseases and what impact things like cosmetic surgery or Botox have on the tool.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
The Sun
a day ago
- The Sun
At-home ‘brain quiz' reveals your risk of heart disease, stroke, dementia and 3 types of cancer
AN at-home 'brain quiz' can reveal your risk of dementia, as well stroke, heart disease and three common types of cancer. It suggests that taking better care of your brain could boost your overall health, researchers said. Developed at Mass General Brigham - a hospital affiliated with Harvard Medical School - the McCance Brain Care Score (BCS) is a tool designed to assess modifiable risk factors that influence brain health. The quiz's 21 questions evaluate someone's physical health, lifestyle, as well as social and emotional factors. All together, these can pain a picture of someone's risk of brain diseases that come on with age. But researchers found the quiz could also shed light someone's risk of non-brain related diseases, such as heart disease or cancer. That's because neurological diseases such as stroke, dementia, and late-life depression, as well as cardiovascular diseases —including ischaemic heart disease, stroke, and heart failure —and cancers are often driven by the same risk factors. These include things like unhealthy diet, lack of exercise, smoking, excessive drinking, high blood pressure, cholesterol and blood sugar. Psycho-social factors like stress and social isolation are also drivers of disease. In fact, at least 80 per cent of cardiovascular disease cases and 50 per cent of cancer cases can be linked to these factors, researchers claimed. Senior author Sanjula Singh, of the McCance Center for Brain Health at Massachusetts General Hospital, said: "While the McCance Brain Care Score was originally developed to address modifiable risk factors for brain diseases, we have also found it's associated with the incidence of cardiovascular disease and common cancers." 'These findings reinforce the idea that brain disease, heart disease, and cancer share common risk factors and that by taking better care of your brain, you may also be supporting the health of your heart and body as a whole simultaneously.' 10 second one leg stand test The quiz will generate a score of 0 to 21 - the higher your score, the more brain-healthy habits you practice. Researchers used data from the UK Biobank to analyse health outcomes in 416,370 people aged 40 to 69 years using the test. They found that a 5-point higher BCS score at baseline was associated with a 43 per cent lower risk of developing cardiovascular disease - including ischaemic heart disease, stroke, and heart failure - over 12 and a half years. For cancer, a 5-point increase in BCS was associated with a 31 per cent lower rate of lung, bowel and breast cancer. How to do the quiz Before you take the test, you'll need some information from your health records, such as your latest blood pressure reading, you haemoglobin A1c score, your cholesterol levels and BMI. The test - available on the Mass General Brigham website - starts off by asking people about their blood pressure, cholesterol levels and BMI, as well as smoking, exercise and sleep habits. It also asks patients about social relationships - whether they're close to anyone outside of spouses and children - and if they feel that their "life has meaning". Once you've answered the questions, the quiz will generate a score between 0 and 21, and ways to improve your brain health. The aim is to achieve the highest score possible. The score will be made up of three categories – physical, lifestyle, and social-emotional health – and it measures what you are already doing to protect your brain and prolong your brain health. This can include exercise, sleep, social interactions, lowering blood pressure and more. "Your McCance Brian Care score is a tool you can use to measure and improve how well you care for your brain throughout your life," according to Mass General Brigham. "Small adjustments, like finding some time to walk more in your day or making it a priority to call your best friend, can have a major improvement on your overall score and brain health over time. "Achieving your highest score and keeping your highest score month after month, year after year, will ensure you are taking the best care of your brain as you grow and thrive." Study authors said that previous research suggests that some individual components of the BCS quiz - such as smoking, lack of exercise and high blood pressure - can raise the risk of dementia or heart disease. But they said that their study can't prove that the BSC components can cause disease - only that they are strongly linked. Researchers also noted that the quiz can't be used to predict whether someone will get a specific disease. Instead, it can serve as a framework to help people identify achievable lifestyle changes that support the health of their brain, as well as the rest of their body. Lead author Jasper Senff said: 'The goal of the McCance Brain Care Score is to empower individuals to take small, meaningful steps toward better brain health. 'Taking better care of your brain by making progress on your Brain Care Score may also be linked to broader health benefits, including a lower likelihood of heart disease and cancer. 'Primary care providers around the world are under growing pressure to manage complex health needs within limited time," he went on. 'A simple, easy-to-use tool like the McCance Brain Care Score holds enormous promise - not only for supporting brain health, but also for helping to address modifiable risk factors for a broader range of chronic diseases in a practical, time-efficient way.'
Daily Mail
30-05-2025
- Daily Mail
Doctors sound alarm over massive spike in Americans suddenly dropping dead from unexplained heart attacks
Americans are dropping dead from heart attacks at home at alarming rates, doctors warn. Researchers at Mass General Brigham, a Harvard-affiliated hospital, found that heart-related deaths gone up to 17 percent in the years following the Covid pandemic. And while the exact number is unknown, in many of these cases heart attacks have been at home, despite deaths in hospitals decreasing. This suggests patients are being 'missed' — until it is too late. Dr Jason H Wasfy, study author and director of Outcomes Research at the at Massachusetts General Hospital, said: 'Lots of reports have shown that there have been fewer heart attacks in hospitals since 2020 — but something seems to be missing from that data. 'We now show that if you account for deaths at home, cardiac deaths are going up and have stayed up for years. Today there are a lot more people having cardiac deaths at home, which also raises the concern that people with heart disease haven't been getting the care they need since the pandemic.' The Covid virus — which is thought to have infected over 100million Americans — has been shown to cause damage to the heart and blood vessels which may be a factor. But doctors say more factors must be at play, which are still being understood — including diet and lifestyle factors. For example, a study this week found smoking marijuana or consuming edibles may raise the risk. Though the full findings have not yet been released, the experts believe the pandemic could also be linked to a mystery rise in young Americans suffering heart attacks. The study, published Friday in JAMA Network Open, looked at 127,746 death certificates from people who died in Massachusetts between January 2014 and July 2024. The average age was 77 and about 52 percent were men. The researchers then made population estimates based on US Census data from 2014 to 2023. Using death certificate and census data, the team set an expected cardiac death rate for 2020 to 2023. They found cardiac deaths were 16 percent higher than expected in 2020, 17 percent higher in 2021 and 2022 and six percent higher in 2023. The amount of monthly cardiac deaths at home also exceeded expected levels between 2020 and 2022 and in hospitals between 2020 and 2023. However, they cited additional data showing heart attack hospitalizations decreased by 20 to 34 percent after the start of the pandemic. This suggests many of these excess deaths occurred at home. The researchers wrote: 'In this population-based cohort study of Massachusetts decedents, we found cardiac deaths increased substantially starting in 2020, with exaggerated seasonal patterns and increases in deaths at home. 'While numerous other studies have found fewer admissions for cardiac emergencies in countries across the world, these studies may have missed events occurring outside of hospitals.' The researchers suggested this increase could be due to more people avoiding hospitals during the pandemic or being dismissed by doctors. CDC research from 2020, for example, found 40 percent of Americans delayed or avoided seeking medical care at the start of the pandemic, and 12 percent stayed away from emergency rooms. However, emerging research also shows the Covid virus itself may cause long-term heart issues, increasing the risk of cardiac deaths. Data shows that heart attack cases are on the rise in young Americans Covid has been linked to myocarditis, inflammation of the heart muscle, and pericarditis, inflammation of the sac-like lining surrounding the heart. With myocarditis, it's thought that Covid causes the immune system to attack itself and cause inflammation of the myocardium, the heart's muscle. This same mechanism has been linked to pericarditis. While most cases are mild, in rare instances, myocarditis can damage the heart and make it difficult for it to pump blood, eventually leading to heart failure, heart attack, and stroke. In cases as rare as one in 200,000, the mRNA in Covid vaccines may trigger a similar immune response and cause myocarditis or pericarditis. Covid-induced inflammation also alters the heart's electrical signals, leading to irregular heartbeats called arrhythmias. Research released this week also pointed to smoking weed or taking edibles at least three times a week as a potential cause of heart attacks. That team found endothelial cells, which line the blood vessels and regulate blood flow, released less nitric oxide in people who regularly smoked marijuana or took edibles. Nitric oxide helps blood vessels dilate and deliver vital oxygen throughout the body. This impaired function affects the blood vessels' ability to dilate, increasing the risk of atherosclerosis, heart attacks, and strokes. Dr John Hsu, senior study author of the new study and director of the Program for Clinical Economics and Policy Analysis at Mass General, said: 'Healthcare systems around the world have experienced multiple shocks since 2020. Our findings suggest that both patient choices about seeking care and outcomes after experiencing a cardiac emergency also have changed. 'Had we not examined mortality using death certificate data, the increases in population cardiac mortality could have gone unnoticed.' There were several limitations to the study, including the lack of data on the causes leading to cardiac death. It's unclear when the full findings will be released. The study was partially funded by the National Institutes of Health.
The Guardian
13-05-2025
- The Guardian
FaceAge: the AI tool that can tell your biological age through one photo
Name: FaceAge. Age: New. Appearance: A computer that predicts how long you'll live. So, it will tell me when I'll die? No thanks. Wait, I haven't even explained it yet. Doesn't matter, it's still the most terrifying thing I've ever heard. No, give it a chance. FaceAge is only doing what doctors already do. Which is what? Visually assessing you to obtain a picture of your health. Oh, that doesn't sound so bad. But FaceAge can do it much more accurately, to the point that it can predict whether or not you'll survive treatment. No, I'm out again. I'll explain more. FaceAge is an AI tool created by scientists at Mass General Brigham in Boston. By looking at a photo of your face, it can determine your biological age as opposed to your chronological age. What does that mean? It means that everyone ages at different speeds. At the age of 50, for example, Paul Rudd had a biological age of 43, according to researchers. But at the same age, fellow actor Wilford Brimley had a biological age of 69. And why does this matter? People with older biological ages are less likely to tolerate an aggressive treatment such as radiotherapy. Repeat all that as if I'm an idiot. OK. The older your face looks, the worse things are for you. Great news for the prematurely grey, then. Actually, no. Things like grey hair and baldness are often red herrings. FaceAge can give a better picture of someone's health by assessing the skin folds on your mouth or the hollowing of your temples. Right, I'll just be off to obsessively scrutinise the state of my temples. No, this is a good thing. A diagnostic tool like this, used properly, could improve the quality of life of millions of people. Although the initial research was confined to cancer patients, scientists plan to test FaceAge with other conditions. I've recently had plastic surgery. Will FaceAge still work on me? Unsure, actually. The creators still need to check that. And what about people of colour? Ah, yes, about that. The model was primarily trained on white faces, so there's no real telling how well it can adapt to other skin tones. This is starting to sound dodgy. Just teething problems. Look how fast AI can improve. Last year, ChatGPT was a useless novelty. Now it's going to destroy almost every labour market on Earth. You'd have to assume that FaceAge will rapidly improve as well. That's reassuring. Yes. Before we know it, it'll be scanning your face and instantly making a chillingly objective judgment call on whether you deserve to live or die. My God, will it? No, of course not. Not yet, anyway. Do say: 'FaceAge is the new frontier of medical diagnostics.' Don't say: 'It says I'm going to die three seconds into the robot uprising of 2028.'
