You can detect biological age and a person's chance of surviving cancer 'with just one photo'
Researchers from Mass General Brigham in Massachusetts have developed an AI tool capable of estimating a person's biological age and predicting their likelihood of surviving cancer using a single facial photograph. The AI was trained on over 40 million images and validated against real patient data. Biological aging refers to the functional state of the body rather than chronological age. The study found that lifestyle factors such as smoking, heavy alcohol consumption, and drug use are strongly correlated with appearing older in photographs. Current smokers were found to look older than former smokers, who in turn looked older than those who had never smoked. The AI's ability to assess health risks based on facial features could potentially aid in personalized medical decision-making, though it is intended to complement, not replace, clinical judgment. The tool is available through a research study website.
A new artificial intelligence tool can estimate a person's biological age and their chances of surviving cancer using a single facial photograph, according to researchers. The technology, developed by scientists at Mass General Brigham in Massachusetts, uses deep learning trained on more than 40 million images to identify health-related indicators. The findings were published in Lancet Digital Health and validated through comparisons with real patient data. Cancer patients, on average, appear approximately five years older biologically than their actual chronological age, compared to individuals without the disease. Those whose faces appeared significantly older than their actual age had worse overall survival rates after cancer treatment. The AI tool, named FaceAge, examines features such as skin quality, texture, wrinkles, bone structure, and fine details like vertical lip lines, eye-area thinning, and changes in facial muscle tone. It excludes common distractions like grey hair or baldness. Hugo Aerts, a professor at Harvard University and director of the AI in medicine programme at Mass General Brigham, stated the tool advances the field of facial health recognition by learning cancer survival risks directly from facial photos. He noted that this is the first instance of an AI model predicting cancer survival from a simple image. The study found that lifestyle factors such as smoking, alcohol consumption, and drug use contribute to visible signs of accelerated aging. Current smokers were found to look roughly three years older than non-smokers. However, former smokers did not show significant differences in facial aging compared to those who had never smoked, indicating potential recovery of skin and facial traits after quitting. Raymond Mak, an associate professor of radiation oncology at Harvard Medical School, highlighted the consistency of the results despite variations in real-world photo conditions, such as lighting, expressions, and makeup. His team aims to enhance the reliability and scalability of facial health assessments to aid personalized medical decisions without replacing professional clinical judgment. FaceAge was trained on data from 58,851 individuals aged 60 or older considered healthy, as well as 6,196 cancer patients from institutions in the Netherlands and the United States. The AI tool has demonstrated the ability to improve survival predictions for patients undergoing palliative care for incurable cancers. Individuals interested in determining their own biological age can participate in the research by uploading a photo via the study website, faceage.bwh.harvard.edu.
Researchers have developed an AI tool capable of analyzing a single facial photograph to estimate a person's biological age and predict their likelihood of surviving cancer treatment. The system uses deep learning trained on over 40 million images to identify health-related indicators such as skin texture, wrinkles, bone structure, and other facial features. Studies show cancer patients tend to appear biologically older than their chronological age, and those appearing significantly older have poorer survival rates. The AI accounts for factors like smoking, which accelerates visible aging, but notes that quitting smoking allows some recovery of facial features. This technology could aid in early detection and personalized healthcare approaches.
Bias read (Center): The article discusses a scientific advancement in medical AI with no direct political implications or controversy. It focuses on technological innovation and health research without taking a stance or showing bias toward any political ideology.
Researchers from Mass General Brigham in Massachusetts have developed an AI tool capable of estimating a person's biological age and predicting their likelihood of surviving cancer using a single facial photograph. The AI was trained on over 40 million images and validated against real patient data. Biological aging refers to the functional state of the body rather than chronological age. The study found that lifestyle factors such as smoking, heavy alcohol consumption, and drug use are strongly correlated with appearing older in photographs. Current smokers were found to look older than former smokers, who in turn looked older than those who had never smoked. The AI's ability to assess health risks based on facial features could potentially aid in personalized medical decision-making, though it is intended to complement, not replace, clinical judgment. The tool is available through a research study website.
Bias read (Center): The article discusses a scientific advancement related to AI and health prediction, with no direct political implications or controversy. It presents findings objectively without apparent ideological framing or bias.
How each side covered it
The same event, grouped by the political lean of the outlets covering it.
progressive
center
conservative
★
How each side covered it
Support independent, bias-aware news and unlock the social pulse, community voting, and every other Supporter feature.
United Kingdom
