A selfie may one day help a doctor judge how resilient a cancer patient is. Researchers at Mass General Brigham in Massachusetts have developed an AI system that estimates a patient's biological age and mortality risk from a single facial photograph, according to sources. The study was published on Monday in The Lancet Digital Health.
The model, FAHR-Face, was trained on more than 40 million facial images and adapted into two tools: FAHR-FaceAge, which estimates biological age, and FAHR-FaceSurvival, which predicts survival directly from the face without relying on age estimates.
The researchers found the survival model focused mainly on areas beneath the eyes and around the nose, while the age model looked at features around the mouth and temples, suggesting the two pick up different health-related patterns.
The survival model was trained on images of 34,363 patients, tested on 3,820 and independently validated on data from 4,892 cancer patients in the Netherlands. Patients in the highest-risk group had more than three times the mortality risk of those in the lowest-risk group, even after accounting for clinical factors.
Patients whose faces looked at least five years older than their actual age had a 21% higher mortality risk, while those looking at least five years younger had a 17% lower risk. Combining the two models improved accuracy further.
The researchers said the technology could help doctors assess patients and guide treatment, but further validation is needed before wide clinical use.