Brief
AI predicts biological age and cancer survival from a single facial photo
A Massachusetts‑based AI model trained on 40 million images can infer health markers, with smoking showing the strongest ageing signal.
By Felo News Desk · Published
Researchers at Mass General Brigham in Massachusetts say an artificial‑intelligence tool can determine a person’s biological age and estimate their chance of surviving cancer from a single facial photograph.
What happened
The model was trained on more than 40 million pictures and then validated against real patient data, according to a study published in Lancet Digital Health. It identified lifestyle influences on facial ageing, noting that heavy alcohol use, drug use and especially smoking make faces appear biologically older.
What the reports add
Hugo Aerts, a professor at Harvard University and director of the AI‑in‑medicine programme at Mass General Brigham, said the work “advances the field of facial health recognition AI a step further by learning survival risk directly from facial photographs of patients with cancer.” Raymond Mak, an associate professor of radiation oncology at Harvard Medical School, highlighted the model’s consistency across variations in pose, lighting, expression and makeup.
What was said
Aerts added, “To our knowledge, this is the first demonstration that an AI model can directly predict cancer survival from a simple photograph.” Mak said, “Our goal is to make face photograph‑based health assessment more quantitative, consistent and scalable so it can support personalised treatment decisions rather than replace physician judgment.”
How it came about
The study follows earlier Felo coverage of AI‑driven ageing research, including a October 2 report on childhood exposures accelerating biological ageing and a October 4 article on AI tools analysing DNA signatures in mouse stem cells.
Key facts
- The AI model was trained on over 40 million facial photographs. (independent.co.uk)
- It can estimate biological age and cancer survival odds from a single photo. (independent.co.uk)
- Smoking showed the strongest correlation with an older‑appearing face. (independent.co.uk)
- The study was published in Lancet Digital Health. (independent.co.uk)
- Researchers invite participants to upload a photo at faceage.bwh.harvard.edu. (independent.co.uk)
Sources
- [1] independent.co.uk — originally reported as “One photo 'can detect a person's biological age and chance of surviving cancer'”
