Development of a smartphone-based multi-spectral imaging system and its performance validation through...

Area of Science:

• Biomedical optics
• Medical imaging
• Dermatology

Background:

• Skin cancers are the most common cancers, necessitating advanced diagnostic and monitoring tools.
• Objective grading of radiodermatitis, a severe radiotherapy side effect, is crucial to prevent treatment cessation.

Purpose of the Study:

• To develop and validate a smartphone-based multi-spectral imaging (MSI) system for skin diagnosis and treatment monitoring.
• To assess the system's performance in evaluating radiodermatitis, specifically erythema.

Main Methods:

• Developed a smartphone-based MSI system integrated with a GPU-accelerated Monte Carlo simulations code.
• Derived skin-layer-resolved physiological parameters (blood volume fraction and oxygen saturation) using Monte Carlo look-up table (MCLUT) methods.
• Utilized analysis of variance (ANOVA) and partial least square discriminant analysis (PLS-DA) for data analysis.

Main Results:

• Achieved a mean square error (MSE) below 0.0031 between derived and experimental spectra, confirming parameter robustness.
• Demonstrated a significant increase (P<0.001) in blood volume fraction and oxygen saturation in erythema via ANOVA.
• Attained 85.9% classification accuracy in distinguishing normal skin from erythema using PLS-DA.

Conclusions:

• The developed smartphone-based MSI system shows significant potential for objective radiodermatitis assessment.
• This technology can aid in monitoring skin cancer treatment and managing radiotherapy side effects.