A research team from Mass General Brigham has published new findings in Nature Communications demonstrating that tracking biological aging through facial photographs can predict cancer survival rates. The study introduces a metric called Face Aging Rate (FAR), which measures the change in a person's estimated biological age over time using artificial intelligence. The results indicate that a higher FAR is significantly associated with decreased survival probability in cancer patients. FaceAge is an AI tool developed by the team that uses deep learning to estimate biological age from a single facial photo. While a 2025 study previously showed that cancer patients often appear about five years older than their chronological age, the current research focused on the utility of analyzing multiple photos taken at different intervals during treatment. The investigators examined photographs of 2,279 patients who received radiation therapy at Brigham and Women's Hospital between 2012 and 2023. FAR was calculated by determining the change in FaceAge between two photos and dividing it by the time elapsed between them. The analysis revealed that patients experienced facial aging 40% faster than their chronological aging. Specifically, accelerated aging, represented by a higher FAR, correlated with lower survival rates. This effect was most pronounced when the interval between photos was two years or longer. The team also calculated FaceAge Deviation (FAD), which estimates how biologically old or young a patient appears relative to their actual age at a single point in time. While patients with both high FAD and FAR values showed poorer survival outcomes, FAR proved more stable and reliable for predicting long-term results compared to single timepoint readings. Raymond Mak, a radiation oncologist and co-senior author of the study, stated that deriving a Face Aging Rate from routine photos allows for near real-time health tracking. He emphasized that measuring biological age over time could refine personalized treatment plans, improve patient counseling, and guide the intensity of follow-up care. Hugo Aerts, director of the Artificial Intelligence in Medicine program, added that tracking FaceAge offers a noninvasive and cost-effective biomarker with potential applications for other chronic diseases and healthy individuals. In a related study published in the JNCI: Journal of the National Cancer Institute, FaceAge was tested on over 24,500 cancer patients aged 60 and older. The study found that those whose biological age was ten or more years older than their chronological age had significantly worse survival outcomes, whereas those within five years of their actual age fared better. The researchers suggest that integrating FAR with baseline FAD could provide a more comprehensive view of a patient's evolving health status. They also noted that future studies must evaluate these metrics in more diverse populations. To facilitate further research, the team has launched an institutional review board-approved web portal allowing the public to submit facial photographs for a FaceAge assessment. This initiative aims to expand the dataset and advance the understanding of how facial aging correlates with health outcomes across various conditions.