AI tool predicts cancer survival from a selfie, study finds

Mass General Brigham system estimates biological age and survival risk from facial photographs

By LineZotpaper

Published19 October 2026

Read Time2 min

Massachusetts General Brigham researchers have developed an artificial intelligence system that can predict a cancer patient's survival odds from a single facial photograph, according to a study published in Lancet Digital Health. The system, trained on more than 40 million images, estimates biological age and directly predicts survival outcomes, potentially offering doctors a low-cost tool to guide treatment decisions.

The AI model, named FAHR (Facial Health Recognition), produces two complementary tools: FAHR-FaceAge, which estimates biological age, and FAHR-FaceSurvival, which directly predicts survival from facial photographs. While FAHR-FaceAge focuses on areas like the creases from the nose to mouth and forehead, FAHR-FaceSurvival evaluates the region below the eyes and around the nasal bridge.

In a study of cancer patients treated with radiotherapy in the United States, those who appeared five or more years older than their actual age had a 21% higher risk of dying over an average four-year follow-up, while those who appeared five years younger had a 17% lower risk. Lifestyle factors, including heavy alcohol use, drug use and smoking, were linked to looking biologically older in the AI's analysis.

"This is the first demonstration that an AI model can directly predict cancer survival from a simple photograph," said Hugo Aerts, co-senior author of the study. Dr Raymond Mak, a radiation oncologist at Mass General Brigham, noted that the goal is to "make face photograph-based health assessment more quantitative, consistent and scalable" to support personalized treatment decisions, not replace physician judgment. The researchers caution that more research is needed before the tool can be widely used.

Analysis

Why This Matters

Could help doctors tailor cancer treatment based on patient's physical reserve and frailty, potentially improving outcomes.
Provides a quick, low-cost assessment compared to traditional tests.
Opens the door for AI facial analysis in broader health assessments.

Background

The idea of estimating biological age from faces has been explored, but this study is the first to directly predict survival from a photo using a large foundation model approach with over 40 million images. The study was published in Lancet Digital Health, a respected journal.

Key Perspectives

Researchers (Hugo Aerts, Raymond Mak): The tool advances facial health recognition AI and could assist treatment decisions without replacing clinical judgment. Potential patients: Could benefit from more personalized care but may have concerns about accuracy and privacy of facial data. Critics/Skeptics: More research is needed to validate across diverse populations and clinical settings. Ethical concerns around AI-based health predictions and potential bias in training data.

What to Watch

Further validation studies in different cancer types and patient populations.
Integration into clinical workflows and regulatory approvals.
Development of privacy safeguards for facial image storage and analysis.

Sources

New AI tool can predict cancer survival by a single patient selfie, study suggests · World news | The Guardian

Checked

12 claims in this story checked against 1 source: 7 supported.

Show each claim

Partly supportedAI tool predicts cancer survival from a selfie, study findsWorld news | The Guardian: “New AI tool can predict cancer survival by a single patient selfie, study suggests”
SupportedMass General Brigham system estimates biological age and survival risk from facial photographsWorld news | The Guardian: “The AI system was custom-built by experts at Mass General Brigham... to estimate biological age and directly predict survival outcomes”
Not in the sourcesModel named FAHR (Facial Health Recognition)World news | The Guardian: “(no expansion of FAHR in source)”
Partly supportedQuote: 'This is the first demonstration that an AI model can directly predict cancer survival from a simple photograph'World news | The Guardian: “To our knowledge, this is the first demonstration that an AI model can directly predict cancer survival from a simple photograph.”
Not in the sourcesQuote: 'make face photograph-based health assessment more quantitative, consistent and scalable'World news | The Guardian: “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.”
SupportedTrained on more than 40 million imagesWorld news | The Guardian: “trained on data from more than 40 million facial photographs”
SupportedThose who appeared five or more years older had a 21% higher risk of dyingWorld news | The Guardian: “the people who... looked five or more years older... had a 21% higher risk of dying during the average four-year follow-up”
Partly supportedThose who appeared five years younger had a 17% lower riskWorld news | The Guardian: “who looked five or more years younger had a 17% lower risk”
SupportedAverage four-year follow-upWorld news | The Guardian: “during the average four-year follow-up”
SupportedLifestyle factors including heavy alcohol use, drug use and smoking linked to looking biologically olderWorld news | The Guardian: “heavy alcohol use, drug use and smoking were linked to looking biologically older in AI analysis of facial photographs”
SupportedStudy published in Lancet Digital HealthWorld news | The Guardian: “published in the Lancet Digital Health journal”
SupportedThe researchers caution that more research is needed before the tool can be widely usedWorld news | The Guardian: “more research was needed before the tool could be widely used”