Mass General Brigham, a leading healthcare provider, has unveiled an AI tool named FaceAge that can estimate a person’s biological age and predict survival odds for cancer patients based on facial images. This deep-learning algorithm, developed by researchers, uses photos to generate predictions of biological age, which reflects the rate of aging compared to chronological age. According to the hospital’s press release, FaceAge was trained on 58,851 images of presumed healthy individuals from public datasets, enabling it to analyze photos of 6,196 cancer patients before radiotherapy treatment.
Among cancer patients, the algorithm found that their biological age was on average about five years higher than their chronological age, potentially signaling a need for more aggressive treatment. The tool was also tested for its capacity to predict the life expectancy of 100 palliative care patients, outperforming the predictions of 10 clinicians. These results were published in The Lancet Digital Health.
“We can use artificial intelligence to estimate a person’s biological age from face pictures, and our study shows that information can be clinically meaningful,” said Hugo Aerts, a key researcher and co-senior author from Mass General Brigham. He emphasized that FaceAge could help doctors personalize treatment decisions and improve care planning, particularly for cancer patients. Dr. Ray Mak, another co-senior author, noted, “This opens the door to a whole new realm of biomarker discovery from photographs, and its potential goes far beyond cancer care.”
However, experts warn that while the technology shows promise, concerns about data bias and ethical implications must be addressed before FaceAge can be integrated into clinical practice. Dr. Harvey Castro, an emergency medicine physician and AI specialist, stressed that AI models are only as reliable as the data they are trained on. “If the training data lacks diversity, we risk producing biased results,” he warned. Castro also highlighted the need for transparency in data ownership, storage, and patient consent, as well as the psychological effects of being told one’s facial image indicates an older biological age without clear context.
Despite these challenges, Castro believes FaceAge could offer valuable insights for oncologists, particularly in prioritizing treatment plans where resilience is key. He called for a balanced approach where AI enhances rather than replaces human judgment, ensuring that technology remains a tool for improvement rather than a replacement for the empathy and context that define medicine.