These 3D images are a composite of two sets of female faces, showing the average facial structure for each age group in the study. The left image shows the average of the 17-29 year-old women, the right 60-77 year-old women. Composite: Chinese Academy of Sciences

Scientists have created a 3D imaging system they claim can reliably predict a person’s biological age from the look of their face alone.

The researchers believe the technology could be used to judge whether proposed anti-ageing treatments have any effect, and to help doctors fine-tune advice and therapies for their patients.

They developed the technology after scanning the faces of more than 300 people aged 17 to 77 and building up a map that reveals how the human face changes over a lifetime.

For each image, the subtle contours of the face were captured in tens of thousands of 3D points in space which the system could then use to assess a person’s biological age. In contrast to chronological age - simply the number of birthdays a person has had - biological age reflects how well (or not) a person’s body is getting older.

From young adulthood to old age, the images captured the process of steady decline. Mouths become wider, noses larger. The corners of the eyes droop. The face loses its smoothness, becomes plump with subcutaneous fat, and the whole lot sags as gravity triumphs over the collagen and elastin fabric of the skin.

The scientist found that most people’s chronological ages lay within six years of their “facial age” as predicted by the computer. So people who had facial ages of 30, were typically aged 24 to 36 years old.

But despite the discrepancy, the scientists found that facial age was a more accurate reflection of a person’s biological age, that is, the state of their bodies as revealed by blood tests that measure signs of ageing.

"I did not expect to see such remarkable changes with age, nor did I expect the 3D images to be such a good biomarker for biological age,” Jing-Dong Han told the Guardian from the Chinese Academy of Sciences in Shanghai. “They turned out to be as accurate as the most accurate marker to date.” What is unclear is whether the same facial features can predict ageing so well in other ethnic groups.

Among the 332 people who took part in the study were a small number who, according to their facial ages, were growing old either much faster or much slower than average. Some of the people’s faces appeared more than ten years older or younger than their real, chronological ages. When Han looked at the blood profiles of these outliers, she found that they too reflected either rapid or much slower ageing than normal.

"The predicted fast agers do have more accelerated ageing blood profiles, and vice versa for the slower agers,” Han said.

She believes that the system, described in Cell Research, could be used to test whether products touted for their anti-ageing effects make any difference. In the longer term, she said, it could aid doctors as a simple way to a measure how well people are ageing, and to help them advise patients on changes to their lifestyles, or to personalise treatments when they fall ill.

"Reliable and easy prediction of the ageing process is not only important for assessing the degree of the ageing process and its reversal, but it is also important for assessing the risks of ageing-associated diseases, for designing personalised treatment schemes, and for individuals to improve their life styles and health,” Han said. (The Guardian)