A selfie can reveal your biological age

Establishing a person's biological age simply from a selfie : it is now possible thanks to a new Artificial Intelligence algorithm , FaceAge , which converts a photo of the face into a number that reflects the age shown instead of that established by the date of birth. Trained on tens of thousands of photographs, this algorithm attributes on average to cancer patients about 5 years more than their chronological age, and could therefore help doctors to establish who is able to withstand aggressive treatments , and who instead requires a more delicate approach. The study was published in the journal The Lancet Digital Health by the group led by the American Mass General Brigham, affiliated with the Harvard School of Medicine.

Aging occurs at different rates for each individual, influenced by a myriad of factors including genes, lifestyle, stress, exercise and habits such as smoking and drinking. Researchers led by Raymond Mak and Hugo Aerts trained FaceAge on nearly 59,000 images of presumably healthy people over the age of 60 , and then tested it with more than 6,000 cancer patients , using photos taken shortly before radiotherapy . The latter show a biological age that is older than their chronological age, and the greater this discrepancy, the more predictive it is of a worse survival rate, especially for those whose biological age is over 85.

FaceAge appears to assess signs of aging differently than humans: for the algorithm, having gray hair or being bald matters much less than subtle changes in facial muscle tone . But this new tool also poses ethical issues : for example, discovering one's biological age could trigger healthy changes but also generate anxiety, or the algorithm could be used by insurance companies and employers against the interests of people. "It's definitely something to pay attention to," says Aerts, "to ensure that these technologies are only used for the benefit of the patient."