Human skin color diversity: computational modeling and Monte Carlo simulations of light transport

• Autor(a) principal: Lima, Victor Porto Gontijo de
• Data de Publicação: 2025
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://www.teses.usp.br/teses/disponiveis/76/76134/tde-03072025-091903/
• topic: Colorimetria; Colorimetry; Cor da pele; Monte Carlo simulations; Refletância espectral; Simulações de Monte Carlo; Skin color; Spectral reflectance

The interaction of light with human skin is fundamental in Biomedical Optics, influencing both diagnostic and therapeutic applications. Skin color varies among individuals and across different body regions of an individual, primarily due to intrinsic differences in the composition and organization of its molecular constituents. The absorption and scattering of light by these molecules determine the fraction of light that escapes the skin and reaches the eye of the observer, ultimately defining perceived skin color. The most influential chromophores are melanin, in the epidermis, and hemoglobin, in the dermis and hypodermis, although others also have their effects. Understanding how these chromophores influence skin color is essential for a better comprehension of skin optics and for optimizing light-based procedures, such as laser treatments and non-invasive diagnostic techniques. This study employed Monte Carlo simulations of light transport to investigate the effects of the volume fractions of melanosomes and blood vessels on skin reflectance and colorimetric parameters (sRGB, CIE L*a*b*, and CIE C*h* coordinates). Changes in the individual typology angle (ITA), melanin index (MI), and erythema index (EI) were also examined. A four-layered model was developed to represent the healthy human skin. Additionally, an open-source Python package, SkinOptics, was created to facilitate computational modeling and data processing. The results demonstrated that an increase in the volume fraction of melanosomes leads to a reduction in reflectance across the entire visible range, resulting in a darkening effect, while the volume fraction of blood vessels primarily affects reflectance between 500 and 600 nm, contributing to skin redness. The results also suggest that incorporating h* alongside ITA may ensure a less misleading skin color classification, particularly in cases in which an erythema is perceptible. Furthermore, a mathematical relationship between MI, EI, and L* was identified, described by the equation of a plane. This relationship was observed in both simulated and experimental datasets and has not been previously reported in the literature. The findings of this work may provide valuable insights into skin optics and colorimetry, with potential applications in dermatological imaging and phototherapy customizations. The developed SkinOptics Python package offers flexible and accessible tools for researchers in the field, enabling further investigations. Future studies should focus on refining skin models, exploring additional chromophores and expanding open access datasets regarding optical properties of differently colored skin.