Facial Image Analysis Tracks Biological Aging, Predicts Cancer Outcomes

In cancer care, clinicians need simple tools that reflect dynamic changes in patient resilience during therapy. Standard measures often miss short‑term shifts that influence outcomes and follow‑up planning. A new study shows that artificial intelligence applied to routine face photographs can track biological aging over time and stratify survival among patients with cancer.

Researchers at Mass General Brigham (Boston, MA, USA) developed FaceAge, a deep learning tool that estimates biological age from a single facial image, and Face Aging Rate (FAR), which quantifies how that estimate changes between time points. The group evaluated whether longitudinal facial aging captured by FAR functions as a noninvasive prognostic biomarker during treatment. The work was conducted across Mass General Brigham programs, including the Mass General Brigham Cancer Institute and the Artificial Intelligence in Medicine (AIM) program.

FaceAge analyzes a patient’s facial features to generate an age estimate that reflects biological rather than chronological status. FAR is calculated as the change in FaceAge between two photographs divided by the interval between them, enabling near real‑time tracking of health trajectory. The team also assessed FaceAge Deviation (FAD), which measures how biologically older or younger a person appears at a single time point relative to chronological age.

In a Nature Communications, investigators analyzed two photographs from each of 2,279 patients with diverse cancers who received at least two courses of radiation therapy at Brigham and Women’s Hospital between 2012 and 2023. Photographs were obtained as part of routine clinical workflow. Higher FAR was significantly associated with decreased survival probability. Median FAR indicated facial aging outpaced chronological aging by 40%, and the association with survival was strongest when photos were spaced two years or more apart. Patients with both high FAD and FAR had poorer survival, while FAR predicted outcomes more stably over longer intervals than FAD.

Complementary evidence from a 2025 JNCI: Journal of the National Cancer Institute study showed FaceAge was older than chronological age in 65% of more than 24,500 radiation therapy patients over 60. Those with FaceAge estimates 10 or more years older had significantly worse survival, whereas estimates five or fewer years older were linked to better outcomes. The authors note that integrating FAR with baseline FAD may provide a nuanced view of evolving health, and prospective studies in more diverse populations are underway.

“Deriving a Face Aging Rate from multiple, routine facial photographs allows for near real-time tracking of an individual's health. Our study suggests that measuring FaceAge over time may refine personalized treatment planning, improve patient counseling, and help guide the frequency and intensity of follow-up in oncology,” said Raymond Mak, M.D., a radiation oncologist at Mass General Brigham Cancer Institute and a faculty member in the Artificial Intelligence in Medicine (AIM) program at Mass General Brigham.

“Tracking FaceAge over time from simple photos offers a noninvasive, cost-effective biomarker with the potential to inform individuals of their health. We hope with continued study we can learn how FaceAge may provide prognostic information for patients with other chronic diseases and for healthy individuals,” said study co-author Hugo Aerts, Ph.D., director of the AIM program at Mass General Brigham.

Related Links
Mass General Brigham