Can a Simple Photo of Your Face Reveal the Severity of Your Illness?
Researchers from Brigham and Women's Hospital are developing AI tools to assess disease severity and biological age by analyzing facial features.
Translated by Qiyas from الراي — Original source
BBC – American researchers say an artificial intelligence tool could help doctors estimate the severity of a patient’s illness based on a single photo of their face.
The scientists explained that the tool, trained on millions of images of healthy faces and 35,000 images of cancer patients, can estimate a person’s resilience to life by closely examining facial features around the eyes.
Although the tool is still far from direct use in hospitals, researchers emphasize that its goal is to assist doctors in recommending appropriate treatments for their patients.
The AI tool, called Face Survival, indicates that the area under the eyes and around the bridge of the nose is the most important part of the face to assess when estimating a patient’s potential lifespan.
These indicators applied to various types of cancer with differing severity levels, including cases where the disease had spread to the patient’s bones or brain.
Most of the cancer patients involved, who were from the United States and the Netherlands, were in their fifties or sixties, but some were children, while others were in their nineties.
The AI tool, for which a patent application is still pending, was designed to assist doctors in this work, not to replace them.
Lead researcher Dr. Raymond Mak, from Brigham and Women’s Hospital in Boston, stated: “As the population ages, doctors need better ways to estimate the body’s resilience and degree of physical frailty before recommending intensive treatments, such as radiation therapy, surgery, or systemic therapy.”
He added: “Our goal is to make face-based health assessments more quantitative, consistent, and scalable, so they support treatment decisions tailored to each patient, rather than replacing the doctor’s judgment.”
Mak and his colleagues are also evaluating an AI tool that predicts biological age—the age a person appears to be—compared to their chronological age since birth.
The Face Age tool indicates that the skin below the nose and near the mouth, specifically the nasolabial folds and marionette lines, along with the forehead (where frown lines appear), the temples (where wrinkles form at the corners of the eyes), and the cheeks, are good indicators of how quickly we age.
Mak said the tool’s results were “strikingly consistent” when applied to common everyday photos, despite variations in facial position, lighting, expressions, and makeup.
He added: “The results remained stable after most cosmetic procedures tested, with only minor changes observed after rhinoplasty.”
However, researchers stress the need for more data and facial images, especially from diverse ethnic groups, to determine the effectiveness of the two tools, which are still in the development stage.
Professor Miroslav Bobek, an expert in video processing and machine learning at the University of Surrey in the UK, said that any claim that the tool can track aging or lifestyle changes over time would be “premature.”
