In the future, computers will be able to guess how old you are
Latest“How old are you?” is one of the most loaded questions on earth. Kids play age-based guessing games at carnivals, adults lie at cocktail parties, and rich people spend millions to stave off the effects of aging. But soon, none of that may matter, because scientists are honing an age-estimation technique that uses 3D-mapping software to figure out how old you are, and—more importantly—how old you look.
Researchers from the Chinese Academy of Sciences-Max Planck Partner Institute for Computational Biology 3D-mapped the faces of more than 330 Chinese people aged 17 to 77. By running these images through an algorithm, the say they’ve generated the “first comprehensive map” of the aging human face. They’re reporting their findings in a paper published today in the journal Cell Research. Some of the results confirm what most olds already know: As we age, our eyes droop. Dark pouches develop under the eyes. The lower jaw widens. Our smile lines deepen. The distance between our nose and mouth lengthens. Our forehead inexplicably narrows. Everything just sags.
The most interesting part of the study came when the scientists tried to tease out whether there are any biological differences between people who look younger than they really are, and those who look older. They started by training their algorithms on the 3D facial morphological data to see how well it did at predicting age. On average, their algorithm came within about six years of guessing participants’ true ages. (Which isn’t great, but, hey, the algorithm is still being a work in progress.)
Using this data as a baseline, the researchers divvied up their subjects by gender and age groups, into slow, fast and normal agers. (They defined “slow agers” as people who were at least six years older than the algorithm guessed. “Fast agers,” on the other hand, were at least six years younger than their algorithmic ages.) Then they looked at biomarkers like cholesterol and albumin, a common blood protein involved in regulating the amount of fluids in blood. Slow-aging women and men, for instance, had more blood albumin than normal or fast-agers. The researchers didn’t delve into how this affected the aging face. But it could be that having just a little bit more albumin helps boost cellular health, and keeps your facial cells from dying and creating facial wrinkles.
The initial sample sizes are small, and the differences between what they’re calling slow and fast agers aren’t that big, so it’s hard to draw any definitive conclusions from this study. But it’s an interesting step toward figuring out how our body’s biology affects our appearance as we age. Once scientists have the personalized records of our changing biomarkers linked to images of our faces throughout our lives, then computers will start to really know how old we are. In the future, lying about your age could well be impossible.
Daniela Hernandez is a senior writer at Fusion. She likes science, robots, pugs, and coffee.