news.today
Hello · Register today | LoginHello · My Account[—] [32]
Tech

AI tool predicts cancer survival from facial photographs

Summarised from 2 outlets · Updated 20 Oct, 03:32 · Archive

Researchers have developed an AI tool that analyses a single facial photograph to estimate a cancer patient's survival chances.

Researchers at Mass General Brigham in Massachusetts have developed an AI tool that analyses a single facial photograph to predict a person's chance of surviving cancer. The tool was trained on more than 40 million images to determine key health indicators and biological age. The cancer survival estimates were then checked against real patient data to verify their accuracy.

Biological ageing refers to how old the body is in terms of health and function rather than the number of years a person has lived. The researchers found that lifestyle factors influence how people age in the face, with heavy alcohol use, drug use and smoking all associated with looking biologically older. Smoking showed the strongest pattern, with smokers appearing older than former smokers, and former smokers appearing older than people who never smoked.

Hugo Aerts, professor at Harvard University and director of the AI in medicine program at Mass General Brigham, said the work 'advances the field of facial health recognition AI a step further by learning survival risk directly from facial photographs'. Raymond Mak, associate professor of radiation oncology at Harvard Medical School, said the team was surprised by how consistent the findings were across variations in pose, lighting, facial expression and makeup. The researchers said their goal is to 'support personalised treatment decisions rather than replace physician judgment'. The study was published in Lancet Digital Health.

AI tool predicts cancer survival from facial photographs
Image: The West Australian

Comments (0)

How it is being reported

In this story: Mass General Brigham · Harvard University · Hugo Aerts · Raymond Mak · Harvard Medical School · Massachusetts · Lancet Digital Health

This summary was written by AI from the headlines and standfirsts above, and states as fact only what at least two outlets report. How we use AI · Report a problem