AI tool predicts cancer survival from a single selfie, study suggests

A new artificial intelligence tool can estimate how long a cancer patient might live using just one photograph of their face, according to a study published in the Lancet Digital Health. The system was custom-built by researchers at Mass General Brigham, a nonprofit academic healthcare system in Massachusetts. It represents one of the most comprehensive demonstrations to date of how AI analysis of a selfie could assess health and help guide medical care.

The technology produced two complementary tools. FAHR-FaceAge estimates a patient's biological age—how old they appear rather than their actual age—and uses that to predict cancer survival. FAHR-FaceSurvival instead predicts survival directly from the photograph. Although both came from the same model, they rely on different parts of the face: the former focuses on creases from the nose to the mouth, the forehead above the eye sockets, and the temples, while the latter examines the area below the eyes and around the nasal bridge.

In a study of cancer patients treated with radiotherapy in the US, those who appeared five or more years older than their actual age had a 21% higher risk of dying during an average four-year follow-up compared with the rest of the group. Patients who looked five or more years younger had a 17% lower risk. “To our knowledge, this is the first demonstration that an AI model can directly predict cancer survival from a simple photograph,” said Hugo Aerts, a co-senior author of the study.

Aerts said the key technical advance was that the model 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 without 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 before recommending intensive treatments as the population ages. He said the goal is to make face photograph-based health assessment more quantitative, consistent, and scalable so it can support personalized treatment decisions rather than replace physician judgment. Researchers also found that heavy alcohol use, drug use, and smoking were linked to looking biologically older in the AI analysis.

Mak noted that FAHR-FaceAge remained consistent across common photo variations such as pose, lighting, facial expression, and makeup. It was also stable across most tested plastic surgery procedures, with only a small shift seen after rhinoplasty. The doctors cautioned that while single facial photographs could offer an easy, low-cost way to support personalized treatment decisions, more research is needed before the tool can be widely used.