Using Monte Carlo simulations to understand the influence of photon propagation on photoacoustic spectroscopic imaging

Abstract

Purpose: The purpose of this study is to evaluate the influence of photon propagation on the NIR spectral features associated with photoacoustic imaging. Introduction: Photoacoustic CT spectroscopy (PCT-S) has the potential to identify molecular properties of tumors while overcoming the limited depth resolution associated with optical imaging modalities (e.g., OCT and DOT). Photoacoustics is based on the fact that biological tissue generates high-frequency acoustic signals due to volume of expansion when irradiated by pulsed light. The amplitude of the acoustic signal is proportional to the optical absorption properties of tissue, which varies with wavelength depending on the molecular makeup of the tissue. To obtain quantifiable information necessitate modeling and correcting for photon and acoustic propagation in tumors. Material and Methods: A Monte Carlo (MC) algorithm based on MCML (Monte Carlo for Multi-Layered edia) has been developed to simulate photon propagation within objects comprised of a series of complex 3D surfaces (Mcml3D). This code has been used to simulate and correct for the optical attenuation of photons in blood, and for subcutaneous tumors with homogenous and radially heterogeneous vascular distributions. Results: The NIR spectra for oxygenated and deoxygenated blood as determined from Monte Carlo simulated photoacoustic data matched measured data, and improving oxygen saturation calculations. Subcutaneous tumors with a homogeneous and radially heterogeneous distribution of blood revealed large variations in photon absorption as a function of the scanner projection angle. For select voxels near the periphery of the tumor, this angular profile between the two different tumors appeared similar. Conclusions: A Monte Carlo code has been successfully developed and used to correct for photon propagation effects in blood phantoms and restoring the integrity of the NIR spectra associated with oxygenated and deoxygenated blood. This code can be used to simulate the influence of intra-tumor heterogeneity on the molecular identification via NIR spectroscopy.