Study finds dorsal hand images as effective as face biometrics for age estimation

Researchers at Haut.AI, a self-described “leader in responsible skincare artificial intelligence (AI) development,” have figured out how to estimate a person’s age using hand biometrics. A release positions their paper on dorsal hand image technology, published in Experimental Dermatology, as a viable alternative to age assurance methods that are dependent on face biometrics.

In “Predicting human chronological age via AI analysis of dorsal hand versus facial images: A study in a cohort of Indian females,”, a team including HautAI’s Anastasia Georgievskaya, Timur Tlyachev, Konstantin Kiselev, Konstantin Chekanov, Daniil Danko and Arseny Golodyaev finds that algorithms trained on hand images are comparable in accuracy to those that use face image, showing an average error of 4.1 and 4.7 years, respectively, in predicting chronological age.

“The correlation of predicted versus actual age is highly statistically significant using either hand or facial image datasets,” says the report. “Thus, dorsal hand images are a viable alternative to facial images for age prediction and provides a potential solution for predicting age in situations where facial images are unavailable or unsuitable, such as in forensic investigations, medical assessments or criminal suspect identification and surveillance.”

The researchers say the study has important ethical implications, particularly considering that its dataset was drawn from the Indian population to include a wide range of skin tones. A key criticism of facial recognition is the bias it has shown against people of color. Anastasia Georgievskaya, the CEO of Haut.AI, says this “has the potential to mitigate biases often associated with conventional systems.”

Beyond algorithmic bias, there are more subtle issues of privacy and identity at play. “Facial images are often considered to be more sensitive than hand images because they are more directly linked to personal identity and can reveal more personal information about an individual, such as age, sex, gender, race, ethnicity and emotions,” reads the report. “Hand images seem to be less sensitive than facial images and may be viewed as more impersonal or less revealing of personal identity.”

A key word, however, is “seem.” The researchers stress that their findings imply not only new opportunities, but also a fresh biometric vulnerability. “The proven possibility of age prediction from hands shows these images can be a source of ethical-sensitive biomarkers. Moreover, it is possible to identify a person based on dorsal hand images as in the case of facial photographs.” In terms of risk, that means a picture of your hand is as chock full of identifiable biometric information as a selfie.

The authors concede that “further research is needed to justify the use of hand images for age prediction.” But, given Haut.AI’s focus on the skincare market, they are bullish on AI’s ability to help understand the aging process. For faces, “researchers found that areas around the eyes, nose, mouth, and forehead were important for facial age prediction by AI; these areas often show wrinkles, sagging, and other signs of aging.” On hands, gnarly knuckles and pruny wrinkles helped estimate age. “Antiaging interventions that address these features,” says the report, “will make you look younger to neural networks and, most likely, to humans, too.”

Article Topics

biometric age estimation  |  biometrics  |  biometrics research  |  face biometrics  |  hand biometrics