Estimation of skin optical properties with a Monte Carlo simulation and a genetic algorithm

Area of Science:

• Biomedical Optics
• Photomedicine
• Computational Modeling

Background:

• Accurate optical properties of skin are crucial for safe and effective laser-based therapies and diagnostics.
• Existing Monte Carlo (MC) models often produce unrealistic spectral outputs due to uniform parameter sweeping, failing to represent diverse skin tones.
• A gap exists in MC databases for simulating realistic skin optical properties across the spectrum of human skin tones.

Purpose of the Study:

• To develop and validate a color-constrained framework for guiding MC simulations of skin optics.
• To improve the accuracy of estimating skin optical properties for different skin tones.
• To bridge the gap between simulated and real skin optical properties for biomedical applications.

Main Methods:

• Implementation of a three-layer skin model in PyXOpto, parameterized by biologically relevant variables.
• Optimization of MC simulations using a genetic algorithm (GA) guided by a color-space constraint.
• Ensuring simulated colors adhere to the individual typology angle (ITA°) bounds for realistic skin tone representation.
• Validation of the framework against ex vivo human skin samples of varying tones.

Main Results:

• The color-constrained framework successfully guided MC simulations to generate spectra within the realistic range of human skin tones.
• The optimized model demonstrated high accuracy in replicating and estimating optical properties of ex vivo human skin across different tones.
• Simulated spectra were shown to be more representative of actual skin compared to traditional uniform parameter sweeping methods.

Conclusions:

• The developed color-constrained framework effectively enhances the accuracy of skin optical property estimation for laser applications.
• This approach provides a more biologically relevant and accurate method for simulating light-tissue interactions in diverse skin types.
• The findings pave the way for improved design and safety of laser-based medical treatments and diagnostics.