Mass General Brigham researchers build AI that estimates cancer survival from facial photos

Summary

Researchers at Mass General Brigham, a healthcare institution in Massachusetts, developed an AI model trained on more than 40 million facial images, called FAHR-Face. FAHR-FaceAge estimates biological age from features such as the forehead, the wrinkles between the nose and mouth, and the temples. FAHR-FaceSurvival was trained on almost 35,000 photos of cancer patients and focuses on the area below the eyes and the bridge of the nose, regions linked with dying. Study author Dr Raymond Mak noted that these are the same features Hippocrates described in 400BC as signs of muscle wasting and mortality risk. Prof Devi Sridhar, chair of global public health at the University of Edinburgh, writes in a Guardian column that she has doubts. She notes that doctors have long used facial appearance as a proxy for general health, but argues that many factors have been ignored and that it is unclear whether AI can reliably predict what will happen to an individual patient.

Timeline

1. The Guardian

   23 Oct 2026, 07:00 UTCSource ↗

   The Guardian reports, in a column by Prof Devi Sridhar, that Mass General Brigham researchers built an AI model on more than 40 million facial images to estimate biological age and predict survival in cancer patients, but Sridhar doubts it can reliably predict individual outcomes because many factors have been ignored.