AI tools predict cancer survival and biological age from a single face photo, study shows

US researchers have published a study showing an AI that predicts cancer survival from a facial photograph and another that estimates biological age.

• Mass General Brigham researchers published their findings in the Lancet Digital Health journal on two tools, FAHR-FaceSurvival and FAHR-FaceAge.
• The system was trained on more than 40 million facial photographs and examined 35,000 photos of cancer patients.
• Dr Raymond Mak said the tools aim to support, not replace, physician judgment and provide scalable health assessment.
• The researchers said the tools are under development and need more data across different ethnic groups before clinical use.
• The study found that cancer patients appearing at least five years older had a 21% higher mortality risk, while those appearing younger had a 17% lower risk.
• The survival-prediction tool estimates a patient's risk by analyzing facial features around the eyes.