A person’s face may reveal more than just emotions; it might offer critical clues about their health. Researchers at Mass General Brigham have developed an AI tool called FaceAge, which uses facial images to estimate a person’s biological age and predict survival outcomes, particularly in cancer patients.
FaceAge was trained on more than 58,000 images of people presumed to be in good health. The researchers then tested the tool on 6,196 cancer patients from two medical centers, using photos taken before they began radiotherapy.
The findings, published in The Lancet Digital Health, suggest the technology could provide valuable insights to guide treatment decisions.
Cancer patients appear older than their actual age
The study demonstrated that cancer patients frequently seemed significantly older than their actual age. On average, their predicted biological age, FaceAge, was five years greater than their chronological age.
Those with older FaceAges had lower survival rates, a trend that remained consistent across different cancer types. The risk was especially high in patients whose predicted age exceeded 85, even when researchers accounted for age, gender, and cancer diagnosis.
Facial data may support better treatment outcomes
“We can use artificial intelligence (AI) to estimate a person’s biological age from face pictures, and our study shows that information can be clinically meaningful,” said Dr. Hugo Aerts, director of the Artificial Intelligence in Medicine program at Mass General Brigham and senior author of the study.
He noted that even a basic photo, such as a selfie, could provide useful health indicators. “How old someone looks compared to their chronological age really matters—individuals with FaceAges that are younger than their chronological ages do significantly better after cancer therapy,” Aerts said.
Visual assessments can be biased and inconsistent
Doctors often rely on visual cues when assessing patients, especially during initial exams. A person’s appearance can offer early signs of vitality or frailty.
Scientists at @MassGenBrigham helped validate FaceAge, a deep learning tool that estimates biological age from face photos and improves survival prediction in cancer patients — showing promise as a future biomarker to guide treatment planning.*@LancetDigitalH pic.twitter.com/yEkFuUZUiJ
— ilyas sahin, MD (@ilyassahinMD) May 9, 2025
However, visual assessments can be subjective. Clinicians may unintentionally let personal impressions of age influence medical decisions. Tools like FaceAge may offer a more consistent, data-driven alternative.
Adding AI to improve prediction accuracy
To explore its usefulness further, the team asked 10 clinicians and researchers to predict short-term survival in 100 patients receiving palliative radiotherapy, based solely on their photos.
Even with access to age and cancer status, their predictions varied and were only slightly better than chance. When FaceAge data was added, their accuracy improved significantly.
Further testing is needed before clinical use
Still, researchers caution that more work is needed before FaceAge can be used in daily clinical settings.
Ongoing studies aim to evaluate the AI tool’s ability to predict cancer outcomes across various hospitals and patient groups, track changes in predicted age over time, and assess its effectiveness when images are modified by surgery or makeup.
Potential for broader health applications
“This opens the door to a whole new realm of biomarker discovery from photographs, and its potential goes far beyond cancer care or predicting age,” said Dr. Ray Mak, co-senior author and faculty member at Mass General Brigham.
“I hope we can ultimately use this technology as an early detection system in a variety of applications, within a strong regulatory and ethical framework, to help save lives.”
