AI Tool Predicts Cancer Survival From Single Selfie, Study Says
A study published in Lancet Digital Health reports that researchers at Mass General Brigham built an AI system that estimates biological age and predicts cancer survival from a single facial photograph. The tool, trained on more than 40 million images, is not yet ready for widespread clinical use.
The model was trained on data from more than 40 million facial photographs across a range of ethnicities and ages. From it, researchers developed two complementary tools. FAHR-FaceAge estimates a patient’s biological age — the age a person appears to be, rather than chronological age — and predicts cancer survival. FAHR-FaceSurvival predicts survival directly from a facial photograph rather than inferring it through biological age.
Although the two tools come from the same model, they capture different information about a patient’s health, the study found. They also rely on distinct parts of the face. FAHR-FaceAge focuses on creases running from the sides of the nose to the corners of the mouth, the forehead area above the eye sockets and the temples. FAHR-FaceSurvival evaluates the region below the eyes and around the nasal bridge.
“FAHR-FaceSurvival advances the field of facial health recognition AI a step further by learning survival risk directly from facial photographs of patients with cancer,” said Hugo Aerts, a co-senior author of the study. “To our knowledge, this is the first demonstration that an AI model can directly predict cancer survival from a simple photograph.”
Among cancer patients in the United States treated with radiotherapy, those who appeared five or more years older than their actual age according to FAHR-FaceAge had a 21% higher risk of dying during an average four-year follow-up than the rest of the group. Patients who appeared five or more years younger had a 17% lower risk.
Aerts said the key technical advance was that the system learned a general representation of the face from more than 40 million images, which could then be adapted to different health outcomes such as biological age and survival. He described the foundation-model approach as a scalable way to develop multiple facial health biomarkers instead of training each model from scratch.
Dr Raymond Mak, a radiation oncologist at Mass General Brigham, said physicians need better ways to estimate physical reserve and frailty as the population ages, particularly before recommending intensive treatments such as radiation, surgery or systemic therapy. “Our goal is to make face photograph-based health assessment more quantitative, consistent and scalable so it can support personalised treatment decisions rather than replace physician judgment,” Mak said.
The researchers also found that lifestyle factors were associated with facial ageing. Heavy alcohol use, drug use and smoking were linked to looking biologically older in the AI analysis. Mak said FAHR-FaceAge remained consistent across common photo variations including pose, lighting, facial expression and makeup, and was stable across most tested plastic surgery procedures, with only a small shift observed after rhinoplasty.
Doctors cautioned that while a single facial photograph could offer an easy, low-cost way to support personalised treatment decisions, more research is needed before the tool can be widely used.
Editor's Summary
The study describes an AI system trained on more than 40 million facial images that can estimate biological age and predict cancer survival from a single photograph. It reported higher mortality risk among radiotherapy patients who appeared older than their actual age, while researchers said more work is needed before clinical use.