Monte Carlo lookup table-based inverse model for extracting optical properties from tissue-simulating phantoms using...

Area of Science:

• Biomedical Optics
• Medical Imaging
• Computational Modeling

Background:

• Accurate measurement of tissue optical properties is crucial for medical diagnostics and treatment planning.
• Existing methods face challenges with complex light-tissue interactions, especially in highly absorbing tissues or with short source-detector distances.
• Developing robust inverse models is essential for reliable optical property extraction.

Purpose of the Study:

• To develop and validate a Monte Carlo lookup table (MCLUT)-based inverse model for extracting optical properties from tissue-simulating phantoms.
• To assess the model's accuracy under conditions of close source-detector separation and high tissue absorption.
• To leverage graphics processing unit (GPU) acceleration for efficient Monte Carlo simulations.

Main Methods:

• Implementation of a Monte Carlo simulation using GPU acceleration to generate a comprehensive lookup table.
• Development of an inverse model based on the generated MCLUT.
• Validation of the inverse model using tissue-simulating phantoms with known optical properties.

Main Results:

• The MCLUT inverse model demonstrated strong agreement between extracted and expected optical properties.
• Achieved low error rates: 1.74% for reduced scattering, 0.74% for absorption, and 2.42% for hemoglobin concentration.
• The model proved effective for scenarios with close source-detector separation and highly absorbing tissues.

Conclusions:

• The MCLUT-based inverse model provides an accurate and efficient method for extracting optical properties from biological tissues.
• This approach is particularly valuable for applications involving superficial tissues or highly vascularized regions.
• The GPU-accelerated Monte Carlo simulation enhances the feasibility of this technique for real-time analysis.