AI Model Developed by Mass General Brigham Predicts Cancer Patient Mortality from Facial Photos

Researchers at Mass General Brigham, a nonprofit academic medical system in Massachusetts, have developed an AI tool that can predict mortality risk in cancer patients from a simple facial photograph. The study was published in The Lancet Digital Health.

The team trained a foundation model called FAHR-Face on more than 40 million facial images. From it, they built two complementary tools: FAHR-FaceAge estimates biological age, while FAHR-FaceSurvival directly predicts cancer patients' mortality risk. The two algorithms rely on different facial regions, with FAHR-FaceAge focusing on nasolabial folds, the forehead and temples, and FAHR-FaceSurvival mainly assessing areas below the eyes and around the nose bridge.

The results showed that AI-estimated biological age is closely related to health status. Among cancer patients receiving radiotherapy, those who appeared at least five years older than their actual age had a 21% higher risk of death during a median follow-up of four years, while those who appeared at least five years younger had a 17% lower risk. Lifestyle factors such as heavy drinking, drug use and smoking were also associated with an older-looking facial appearance.

Co-author Hugo Aerts said this is the first study to show that an AI model can directly predict cancer patients' survival prognosis from a single photo. The researchers said the tool aims to make facial-photo-based health assessment more quantitative, standardized and scalable, assisting personalized treatment planning rather than replacing physician judgment.