AI analyzes single facial photo to estimate biological age and cancer survival odds
Researchers have created an AI system that can infer a person’s biological age and likelihood of surviving cancer from one facial photograph.
A team from Mass General Brigham developed an artificial-intelligence model that reads a single face picture to estimate biological age and predict cancer survival probability. Using a dataset of over 40 million photos, the system learned how lifestyle factors like smoking, heavy drinking and drug use manifest as facial aging, with current smokers appearing oldest. The researchers cross-checked the cancer-survival predictions against actual patient outcomes, confirming the model’s reliability.
Hugo Aerts described the work as a step forward for facial health-recognition AI, while Raymond Mak noted its consistency across variations in pose, lighting and makeup. The authors invite the public to test the algorithm via a study website, aiming to make health assessment more quantitative and scalable.
Why it matters
It shows how AI could turn everyday photos into a quick, non-invasive health screening tool.
In this story
