Can AI predict cancer survival from a selfie, new study explains

Researchers have developed an AI model that estimates biological age and cancer survival risk using a facial photograph. The technology could eventually help doctors assess patients before cancer treatment, but experts need to validate its accuracy and safety across different populations.AI tool uses facial photos to estimate biological age and cancer survivalNew Delhi,Oct 21, 2026 09:55 ISTA photograph of a person's face may reveal more about their health than previously understood. Researchers at Mass General Brigham, a healthcare system in the United States, have developed an artificial intelligence (AI) model that can estimate biological age and predict cancer survival outcomes using a single facial photograph. The findings, published in The Lancet Digital Health, suggest that facial features may contain clues about a person's underlying health and physical resilience.The AI system, called Foundation Artificial Intelligence for Health Recognition, or FAHR-Face, was trained on more than 40 million facial images. Researchers developed two tools using the model. One estimates biological age, while the other assesses cancer survival risk directly.The findings could be relevant to personalised cancer care, where doctors consider a patient's overall health and ability to tolerate treatment. However, the technology is still being evaluated. It cannot diagnose cancer, determine an individual's life expectancy or replace a medical assessment.For India, where cancer care takes place across different healthcare settings, the idea of using a simple photograph to gather additional health information may be promising. But the model's performance must be tested among Indian patients before it can be considered for clinical use. HOW DOES THE AI TOOL PREDICT CANCER SURVIVAL FROM A FACE PHOTOFAHR-Face uses deep learning, a type of AI that identifies patterns in large amounts of data. Researchers developed two complementary tools to analyse facial features.FAHR-FaceAge estimates biological age, which refers to how old the body appears to be functioning compared with a person's actual age. FAHR-FaceSurvival predicts cancer survival outcomes directly from a photograph. The tools analyse different parts of the face. The biological age model examines features such as the forehead, temples and creases running from the nose towards the corners of the mouth. The survival model focuses more on the area beneath the eyes and around the bridge of the nose.Researchers have not established which biological processes explain these associations. The model identifies statistical patterns in photographs, rather than detecting cancer or measuring a person's exact chances of survival.WHAT DID THE STUDY FIND ABOUT BIOLOGICAL AGE AND CANCER SURVIVALResearchers found an association between AI-estimated biological age and mortality among cancer patients who received radiotherapy in the United States.During an average follow-up of about four years, patients whose estimated facial age was at least five years older than their actual age had a 21 per cent higher risk of death than the comparison group. Patients who appeared at least five years younger had a 17 per cent lower risk.These figures describe differences between groups. They do not mean that an individual patient has a 21 per cent probability of dying or that looking younger guarantees better survival.The researchers also found that smoking, heavy alcohol consumption and drug use were associated with appearing biologically older in the AI analysis. These findings do not prove that facial ageing causes poorer cancer outcomes.HOW COULD AI-BASED CANCER PROGNOSIS HELP DOCTORSDoctors consider a patient's overall health when deciding on cancer treatment. Surgery, radiotherapy and medicines such as chemotherapy can affect patients differently, depending on their physical condition, existing illnesses and ability to recover.The researchers believe facial analysis could eventually provide additional information during these assessments.Dr Raymond Mak, a radiation oncologist at Mass General Brigham Cancer Institute and a corresponding author of the study, said the aim was to make photograph-based health assessments more consistent and scalable, so they could support personalised treatment decisions rather than replace doctors.If future studies establish the model's reliability, it could help identify patients who need a closer assessment of their physical condition. However, doctors would still need to consider cancer type, stage, medical history, test results, imaging and patient preferences.WHAT COULD THE FINDINGS MEAN FOR CANCER PATIENTS IN INDIAIndia has a substantial cancer burden, making improvements in treatment planning and access to appropriate care important public health priorities. A tool that requires only a photograph could eventually offer an additional way to assess patients.However, a model trained on data from other populations may not perform equally well among Indian patients. Facial characteristics, demographic profiles, disease patterns and healthcare experiences can differ.Researchers would need to validate FAHR-Face across diverse Indian populations, cancer types and treatment settings. They would also need to establish whether it improves predictions beyond existing clinical assessments.Privacy is another concern. Facial photographs are sensitive personal data, and any use in healthcare would require informed consent, safeguards and transparency about how the information is collected and used.IS THE AI TOOL READY FOR CLINICAL USENo. The researchers have said further work is needed before facial photographs can be used widely to guide cancer treatment decisions.An AI model may identify patterns linked to survival without explaining why those patterns exist. Researchers must test it in independent populations and determine whether its predictions improve patient outcomes.Patients should not interpret their facial appearance or an AI-generated biological age estimate as a reliable measure of their individual cancer prognosis.The study offers a possible new direction for personalised cancer assessment. But for India, the next step is rigorous local validation and safeguards against bias and misuse. A selfie may eventually provide doctors with an additional piece of information, but it cannot replace a patient's medical history, clinical examination or a doctor's judgement.- Ends