Health · Medical Research · 1 hr ago
AI facial-photo model links biological age with cancer survival risk
Researchers developed AI models that use facial photographs to estimate biological age and assess survival risk in cancer patients.
One model, FAHR-Face, was trained on more than 40 million face photos and estimated biological age with an average error of about five years.
In a group of 38,183 patients treated at six US centres, those who appeared biologically older than their actual age tended to have poorer survival outcomes.
The highest-risk group was more than three times as likely to die as the lowest-risk group in a separate test of 3,820 cancer patients, even after researchers accounted for other factors.
Tests involving 4,892 patients in the Netherlands also suggested the models could help estimate risk outside the original US group.
The models cannot diagnose cancer or predict an individual’s lifespan, and photo quality, lighting and facial expressions may affect their estimates.
Further testing is needed before routine clinical use, including to establish how well the technology works in populations such as India.
Researchers developed an artificial intelligence model that uses facial photographs to estimate biological age and cancer survival risk.
Patients in the highest-risk group were more than three times as likely to die as those in the lowest-risk group, the study found.
The model was tested on cancer patients in the United States and the Netherlands.
A Delhi oncologist said its reliability in India remains to be established.
The researchers said further validation is needed before routine clinical use.
- Who
- Researchers developed and tested the models; patients with cancer were included in the study.
- What
- The models use facial photographs to estimate biological age and survival risk.
- When
- The findings were published in 2026.
- Where
- The models were tested on patients treated at centres in the United States and the Netherlands.
- Why
- The models were developed to estimate biological age and assess cancer survival risk.
This story does not have two clearly opposing sides.
FAHR-Face was trained on more than 40 million facial photographs.
Researchers assessed survival risk using photographs from 34,363 cancer patients and tested the model on another 3,820.
Researchers tested the models on 4,892 cancer patients treated in the Netherlands.
The findings were published in The Lancet Digital Health.
- Training photographs
- More than 40 million
- US patient group
- 38,183 patients treated at six centres affiliated with Harvard
- Survival model training group
- 34,363 cancer patients
- Survival model test group
- 3,820 cancer patients
- Netherlands test group
- 4,892 cancer patients
1 source
What people are saying
An important consideration is whether this facial recognition AI can deliver equally reliable results in the Indian population, given our distinct epidemiological profile and diverse facial features. This remains to be established,
Rather than viewing it as a standalone technique, its potential may lie in complementing established diagnostic approaches, including genomics and liquid biopsy, as part of a more comprehensive, personalised approach to cancer care,