Establishment of a PCA model for skin health evaluation

摘要

The aim of this study is to establish an objective skin evaluation system through testing skin data and scoring conducted by general dermatology experts. First, 18 data points including left cheek face water evaporation, and left cheek oil content are collected. Second, comprehensive skin information is obtained from volunteers by the mathematical model of principal component analysis (PCA), following scoring the status of the volunteers’ skin by dermatology experts. Then, 99 volunteers are randomly divided into two groups, namely model information and test information. Specifically, skin model information is obtained from 80 volunteers, and test information is obtained from 19 volunteers. PCA is used to extract 18 independent variables of skin test data by means of R Studio software, followed by principal component regression by expert scoring and the extracted principal components. Finally, based on the expert score and model score, skin grading classification can be conducted by MATLAB software, in order to intuitively and effectively evaluate skin status.