Artificial intelligence may one day help doctors assess a patient’s health by analysing a simple photograph of their face. Researchers in the United States have developed an experimental tool called FaceSurvival that examines facial features to estimate survival outcomes among cancer patients.
The findings, published in The Lancet Digital Health, suggest that facial images could offer additional information to support medical assessments. However, researchers stress that the technology requires further testing before it can be used in hospitals.
How Does FaceSurvival Work?
The AI tool was trained using millions of photographs of healthy faces and approximately 35,000 images of cancer patients. It analyses facial patterns to estimate survival prospects across different cancer types and levels of disease severity.
Researchers found that areas around the eyes and the bridge of the nose were particularly important to its predictions.
The patients included in the research came from the United States and the Netherlands and represented a range of ages and cancer conditions.
Can Facial Features Reveal Health Risks?
Doctors already assess patients through physical examinations, medical histories and diagnostic tests to understand their overall condition and likely treatment outcomes.
FaceSurvival is being developed as a potential additional source of information. Researchers hope it could help doctors evaluate frailty and physical resilience when considering demanding treatments such as surgery, radiation therapy or systemic cancer treatment.
However, a facial photograph cannot independently establish a diagnosis or reliably determine an individual’s future health.
Another AI Tool Estimates Biological Age
The research team is also developing FaceAge, an AI tool designed to estimate biological age rather than chronological age.
The system analyses facial features, including forehead lines, wrinkles around the eyes, changes near the mouth and the appearance of the cheeks. Researchers are investigating whether these features can provide clues about the ageing process.
Lifestyle factors such as smoking and heavy alcohol consumption can affect how a person’s skin and face age. Nevertheless, facial appearance alone cannot provide a complete picture of a person’s biological health.
Why More Research Is Needed
Experts say the technology remains at an early stage. Researchers need to test its performance across more hospitals, patient populations and ethnic groups to determine whether its predictions are reliable and fair.
Differences in lighting, facial expressions, makeup and camera angles can also affect photographs, even when an AI system appears relatively consistent under tested conditions.
Another important question is whether the tool can explain why it predicts a particular outcome. Making a prediction does not necessarily reveal the underlying medical cause.
Can It Be Used in Hospitals Now?
Not yet. FaceSurvival and FaceAge remain experimental tools, and their results should not be used as substitutes for medical examinations or professional advice.
If further research establishes their accuracy and safety, such systems could eventually help doctors gather additional information when planning personalised care. For now, their potential remains promising but unproven in routine clinical practice.
