Researchers have developed an AI model that can not only estimate biological age from simple facial photos but also identify health risks, including mortality risk in cancer patients. Below, we will detail the practical effects of this technology. Additionally, there is a dedicated research website where you can experience it yourself.
Recently, an interesting study was published in The Lancet Digital Health. In this study, researchers developed an AI model called FAHR-Face (Foundation for AI Health Recognition). The model was trained on over 40 million facial images and fine-tuned for two key tasks. These two models capture different types of information:
The first AI model estimates a person's biological age by analyzing facial features. This may differ from their actual age but helps assess their overall health status and physical resilience.
The second model estimates the survival risk of cancer patients based on facial photos. The areas under the eyes and along the bridge of the nose appear to be particularly important for making these predictions.
How reliable are the results?
Overall, researchers were granted access to data from approximately 43,000 cancer patients to evaluate the AI model. They then conducted another test on 3,820 of these patients. The model classified 122 of them as high-risk. Among these patients, 52% died within 90 days.
The model classified 975 patients as the lowest-risk group. During the same period, only 2% of these patients died.
Significance
A major advantage of this approach is its low technical and economic barrier. Anyone can take a simple facial photo in seconds.
However, a clear causal relationship has not yet been established. AI can identify statistical patterns but cannot automatically explain the biological processes leading to these patterns. Furthermore, it is unclear whether these findings apply to patients with other diseases.
Conclusion
Artificial intelligence appears to be able to extract information related to biological age and mortality risk in cancer patients from a single facial photograph. Studies show that this association persists even when clinical prognostic factors are taken into account. However, it remains unclear how much additional predictive value facial photographs provide compared to existing clinical methods.
Anyone who wants to try using facial AI to estimate their own biological age canRegister on this website and participate in this "open study". Users can upload 3 to 10 facial photos, either all at once or daily. The system provides an individual biological age score for each photo, as well as a total score. Privacy is protected: photos are not shared and are automatically deleted after a certain period. Other information is anonymized and made available to other researchers.
