Artificial intelligence developed to assess cancer patients' survival probability

Researchers working at the Mass General Brigham healthcare organization in the US have developed a new artificial intelligence model called FAHR-Face, which can assess cancer patients' survival probabilities from a single facial photograph.

A new artificial intelligence system developed in the US can make important predictions regarding the health status of cancer patients from a single facial photograph. According to researchers, this method could also assist in treatment decisions in the future. The new research, reported by the BBC from the journal The Lancet Digital Health, revealed that facial features may carry important information not only about aging, but also about a person's general health status. According to scientists, this technology can help evaluate the physical condition of patients, especially before intensive treatments.

Researchers trained the artificial intelligence model named FAHR-Face with more than 40 million facial photographs belonging to individuals from different age groups and ethnic backgrounds. Based on this model, two separate systems were created. The first tool, named FAHR-FaceAge, estimates a person's biological age by examining signs of aging on the face. The second tool, named FAHR-FaceSurvival, directly evaluates the survival probabilities of cancer patients from their facial photographs.

As a result of the research, it was determined that cancer patients whose biological age was calculated to be at least five years older than their chronological age had a 21 percent higher risk of death, while patients whose biological age was calculated to be at least five years younger than their chronological age had a 17 percent lower risk of death. Researchers stated that further scientific studies are needed for the system to be used in hospitals.