Scientists at Mass General Brigham, a Massachusetts-based academic healthcare system, have created an artificial intelligence tool designed to estimate biological age and forecast survival outcomes for cancer patients using facial photographs. The technology represents one of the most extensive applications of AI facial analysis for health assessment purposes and was developed using training data from more than 40 million facial images representing diverse ethnicities and age groups. Findings from the research were published in the Lancet Digital Health journal.
The researchers developed two complementary systems from the same underlying model. FAHR-FaceAge estimates a patient’s biological age—the age a person appears visually—and predicts cancer survival rates, while FAHR-FaceSurvival forecasts survival probability directly from facial analysis without calculating biological age as an intermediate step. Despite originating from identical technology, the two tools utilize different facial regions to generate their respective predictions. FAHR-FaceAge examines nasolabial folds, forehead creases, and temple areas, whereas FAHR-FaceSurvival focuses on periorbital and nasal bridge regions.
In a study examining cancer patients who received radiotherapy, those appearing five or more years older than their chronological age demonstrated a 21 percent elevated mortality risk over an average four-year monitoring period. Conversely, patients appearing five or more years younger showed a 17 percent reduction in mortality risk. Researchers noted that lifestyle factors including alcohol consumption, substance use, and smoking correlated with appearing biologically older in the AI analysis.
Mass General Brigham researchers emphasized that while facial photograph analysis offers a potentially simple and inexpensive method for supporting personalized treatment decisions, additional investigation is needed before widespread clinical adoption. The tool could assist physicians in assessing patient frailty and physical reserve before recommending intensive treatments such as radiation therapy, surgery, or systemic therapy. Notably, the AI system demonstrated consistency across typical photographic variations including lighting, pose, and facial expression, with minimal changes observed following most cosmetic procedures except rhinoplasty.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI