AI facial analysis estimates cancer patients' survival risk

An AI model trained on over 40 million facial photos can estimate survival risk in cancer patients, with those in the highest-risk group showing more than three times the mortality rate of the lowest-risk group. Researchers at Mass General Brigham developed FAHR-FaceSurvival, fine-tuned on 34,363 patient photos, and validated its findings in a separate group. A second tool, FAHR-FaceAge, estimates biological age, whose gap from actual age was independently linked to survival. The two AI tools focus on different facial regions, suggesting they capture distinct health signals. Researchers emphasize the technology is meant to support, not replace, doctors' judgment in personalized treatment decisions.