News

News

07/12/2023 CosmeticBusiness

AI Identification of Skin Care Needs Is on the Nose

Photographs Used to Identify Signs of Future Ageing

Researchers at Japanese cosmetics company Shiseido have developed a "beauty algorithm" designed to predict the occurrence of signs of ageing with the help of images. According to the company, the team has succeeded in evaluating the relationship between the nasal skeleton structure and the characteristics of tissues and capillaries and therefore the future emergence of wrinkles or sagging skin. The algorithm the experts have developed on the basis of these insights aims to allow consumers to adapt their skin care routine to expected changes as a preventive measure and recommend suitable steps to them.

Targeting Skin Problems Before They Appear

The company had previously developed methods and technologies that involved using images to evaluate skin condition and suggest an individual skincare regime. They not only looked at skin characteristics that a person is born with but also the impact of lifestyle. The manufacturer's research now focuses on addressing skin problems before they even appear.

Gaining a Better Understanding of Existing Skin Problems

The researchers analysed facial images obtained from 424 Asian women aged 40–59. They found that skin conditions such as wrinkles and sagging skin are strongly related to the features of the nose. The root, bridge and wings of the nose played a particularly strong role. The findings fed into the development of an image processing technology that is able to determine and evaluate the features of individuals' nasal structure using smartphone cameras. A tool was created that can classify faces into 16 types and predict skin problems that are likely to appear in the future for each. The company states that the aim of the beauty algorithm is not only to enable preventive skin care but also to better understand problems that have already appeared on the skin and address them in a more targeted way.

Source: Shiseido

Adobe Stock, Leegudim
Back to all news