基于皮肤本态数据库的中青年女性面部皮肤综合状态评价模型构建

Construction of a comprehensive evaluation model for the facial skin of young and middle-aged women based on the Chinese female skin database

从中国女性皮肤本态数据库(北京工商大学)中随机抽取22~42岁女性志愿者1 085例,采集志愿者的皮肤数据,通过相关性分析初步降噪,得到皮肤状态核心指标,并采用KMO检验和巴利特球形检验方法进行检验。根据主成分分析法计算评价模型的各项系数,最终构建皮肤状态综合评价模型。皮肤状态相关的核心指标包括:年龄、皮肤总弹性、皮肤光泽度、黑色素含量、血红素含量、水分含量、经皮水分散失量、肤色ITA°、油脂含量、毛孔数量等10项。该模型对女性的综合皮肤状态给予客观的量化判定,弥补了肉眼判断受到的周围环境以及个人主观意识的影响,解决了皮肤评价的时空限制,采集相关指标数据即可对女性皮肤进行有效评价。研究结果可为女性皮肤状态综合判定提供技术支持。

The changes in the appearance of facial skin are an obvious physiological phenomenon in the process of human aging. People are increasingly concerned about the comprehensive skin status. In recent years, newly developed instruments or methods have continuously emerged in the field of non-invasive skin testing. Skin testing and skin photography are the most widely used. Through the analysis of skin testing and skin photography data, objective and true skin data can be obtained to quantify the skin status. Therefore, it is necessary and feasible to establish a new comprehensive skin condition evaluation model to achieve objective and accurate quantification of skin status. Herein, 1,085 female volunteers aged 22-42 were randomly selected from the Chinese Female Skin Testing Database(Beijing Technology and Business University). The data of skin physiological indicators, VISIA front face and left face crosspolarized light image data(extracted by machine learning algorithms) were collected for correlation analysis between indicators and principal component analysis(PCA) to obtain the core indicators for evaluating individual skin status. KMO test and Bartlett’s test sphere test were used to test the reliability of the principal component data, and the coefficients of each item in the evaluation model were calculated by the method of principal component analysis. There were 10 core indicators related to skin conditions, including age, skin elasticity, skin glossiness, melanin content, heme content, skin moisture content, transdermal water loss, ITA°, sebum secretion, and pores. Eventually, a comprehensive skin status evaluation model was constructed, and the quantified skin score could be obtained by inputting the data of skin physiological indicators into the model. This model gives an objective quantitative judgment on the comprehensive skin condition of women, which avoids the influences of the surrounding environment and the personal subjective consciousness from the naked eye judgment and solves the limitation of time and space for skin evaluation. Collecting relevant index data can be effective on women’s skin evaluation. The research results can provide technical support for comprehensive evaluation of women’s skin status.