AI tool estimates cancer survival risk from a facial photograph

Researchers at Mass General Brigham have developed an artificial intelligence system that estimates biological age and cancer survival risk using a single facial photograph, according to a study published in *The Lancet Digital Health*.

The system, called FAHR-Face, was built from a model trained on more than 40 million facial images spanning a range of ages and ethnicities. Researchers adapted it into two complementary tools: FAHR-FaceAge, which estimates how old a person appears biologically, and FAHR-FaceSurvival, which predicts survival risk directly from the photograph rather than through an age estimate.

The tools examined different facial regions. FAHR-FaceAge focused largely on areas around the mouth, forehead and temples, while FAHR-FaceSurvival relied more on areas beneath the eyes and around the bridge of the nose.

The survival model was trained using images from 34,363 patients and tested on 3,820 others, according to The Express Tribune. It was also independently validated with data from 4,892 cancer patients in the Netherlands. Patients placed in the highest-risk group had more than three times the mortality risk of those in the lowest-risk group after clinical factors were taken into account.

Among U.S. cancer patients receiving radiotherapy, those whose estimated biological age was at least five years above their chronological age had a 21% higher mortality risk during an average four-year follow-up. Those estimated to look at least five years younger had a 17% lower risk. Researchers said further validation is required before the technology can be widely used in clinical practice.