Acquisition of skin characteristics by Monte Carlo modeling and evolutionary setting of parameters

• Published in: Skin research and technology Vol. 26; no. 5; pp. 740 - 748
• Main Authors: Cantú Rodríguez, José Mario, Puente Ramírez, Norma Patricia, Montes Tapia, Fernando Félix, Torres Treviño, Luis Martín
• Format: Journal Article
• Language: English
• Published: England John Wiley & Sons, Inc 01.09.2020
• Subjects: Algorithms; Computer applications; Computer simulation; Dermis; diffuse reflectance; Epidermis; evolutionary algorithm; Evolutionary algorithms; Genetic algorithms; Hemoglobin; Hypodermis; Mathematical models; Monte Carlo method; Monte Carlo simulation; Optimization; Order parameters; Reflectance; Robustness (mathematics); Skin; Spectrophotometry; Thickness; skin; Monte Carlo method; diffuse reflectance; evolutionary algorithm
• ISSN: 0909-752X; 1600-0846
• DOI: 10.1111/srt.12866

Background In this paper, we present the intelligent system to characterize optically human skin; our proposal is a non‐invasive way to obtain some parameters of the skin such as the concentration of hemoglobin, water percentages, and thickness of the layers of the skin. Material and methods To achieve the objective of this work, we used an experimental technique called diffuse reflectance spectrophotometry and numerical calculations, such as the Monte Carlo method and the evolutionary algorithm Evonorm. Results Five case studies were performed. In the first two cases with the Monte Carlo method, a simulated diffuse reflectance was obtained with proposed parameters in order to compare the parameters obtained by the evolutionary algorithm and the proposed parameters. In the rest of the cases, an experimental diffuse reflectance obtained from volunteers was used. Conclusions Numerical modeling was presented to non‐invasively detect some parameters of the skin such as hemoglobin concentration, water percentages, and the thickness of the epidermis, dermis, and hypodermis. It was proposed to use evolutionary algorithms for being robust methods for the optimization of complex problems with a reasonable computational cost.