AI Tool Can Predict Cancer Survival from a Selfie

By the Editorial Team · Friday, October 23, 2026

Researchers at Mass General Brigham have created an AI tool that predicts cancer patients' survival chances by analyzing their facial photographs to estimate biological age. This method, which identifies patients who appear older or younger than their actual age, could enhance personalized treatment planning by providing a quick, noninvasive way to assess health risks alongside traditional clinical evaluations.

Researchers at Mass General Brigham in Massachusetts have developed an artificial intelligence (AI) tool capable of predicting cancer patients’ survival probabilities based on photographs of their faces, according to a study published in the journal Lancet Digital Health. The AI system assesses the biological age of patients by analyzing their facial images and comparing the estimated age to their actual chronological age.

The study focused on cancer patients in the United States who received radiotherapy treatment. Results showed that individuals who appeared to be five or more years older than their true age had a 21% increased risk of mortality during an average follow-up period of four years compared to others in the cohort. Conversely, those who were identified by the AI as looking five or more years younger faced a 17% lower risk of death within the same timeframe.

These findings highlight a notable correlation between biological age as estimated through facial analysis and overall physical health, suggesting that visible signs of aging can serve as indicators of a patient’s underlying health status. The researchers propose that this AI-based approach could serve as a noninvasive tool to supplement traditional clinical assessments in oncology settings, potentially aiding in personalized treatment planning and risk stratification.

The study underscores the potential of machine learning and computer vision technologies in enhancing medical prognostics, particularly by providing rapid, low-cost assessments that do not require invasive procedures. However, further research and validation across diverse populations and cancer types are necessary to confirm the tool’s broader applicability and accuracy.

While the technology presents promising implications, experts caution that it should be integrated carefully alongside established diagnostic methods rather than serving as a sole predictor of patient outcomes. As this AI system matures, it may contribute to improving survival predictions and ultimately support better clinical decision-making for cancer patients worldwide.

This article was reported by synthesizing coverage from multiple independent news outlets to present a balanced, unbiased account. Read about our editorial standards. Spotted an error? Let us know.