Monte Carlo (MC) simulation was implemented in a three dimensional tooth model to simulate the light propagation in
the tooth for antibiotic photodynamic therapy and other laser therapy. The goal of this research is to estimate the light
energy deposition in the target region of tooth with given light source information, tooth optical properties and tooth
structure. Two use cases were presented to demonstrate the practical application of this model. One case was comparing
the isotropic point source and narrow beam dosage distribution and the other case was comparing different incident
points for the same light source. This model will help the doctor for PDT design in the tooth.
The ability to separately measure the scattering coefficient (μ<sub>s</sub> [cm<sup>-1</sup>]) and the anisotropy (g) is difficult, especially
when measuring an in vivo site that can not be excised for bench-top measurements. The scattering properties (μ<sub>s</sub> and g)
can characterize the ultrastructure of a biological tissue (nuclear size, mitochondra, cytoskeletion, collagen fibers,
density of membranes) without needing an added contrast agent. This report describes the use of reflectance-mode
confocal scanning laser microscopy (rCSLM) to measure optical properties. rCSLM is the same as optical coherence
tomography (OCT) when the OCT is conducted in focus-tracking mode. The experimental measurement involves
translating the depth of focus, z<sub>f</sub>, of an objective lens, down into a tissue. As depth z increases, the reflected signal R
decreases due to attenuation by the tissue scattering (and absorption, μ<sub>a</sub>). The experimental data behaves as a simple
R(z) = ρ exp(-μz<sub>f</sub>)
where ρ is the local reflectivity (dimensionless) and μ [cm<sup>-1</sup>] is an attenuation coefficient. The relationship between
(ρ,μ) and (μ<sub>s</sub>,g) is:
μ = (μ<sub>s</sub> a(g) + μ<sub>a</sub>) 2 G(g,NA)
ρ = μ<sub>s</sub> L<sub>f</sub> b(g,NA)
where a(g) is a factor that drops from 1 to 0 as g increases from 0 to 1 (determined by Monte Carlo simulations)
allowing photons to reach the focus despite scattering, G is a geometry factor describing the average photon pathlength
that depends on the numerical aperture (NA) of the lens and the anisotropy (g), L<sub>f</sub> is the axial extent of the focus, and
b(g,NA) is the fraction of scattered light that backscatters into the lens for detection.
Preparation of phantoms with reproducible and homogenous optical properties is tricky. The microscopic
heterogeneity and macroscopic homogeneity of tissue phantoms were compared using reflectance-mode
confocal laser scanning microscopy. Tissue phantoms were prepared using polystyrene microspheres as
scattering medium in aqueous and gel matrix. Uniform distribution of microparticles in phantoms was
evaluated by confocal imaging. Comparison of the heterogeneity of the phantoms was accomplished based
on microscopic optical scattering properties. Distribution of optical properties at the microscopic levels was
determined by a simple theory developed based on the depth-dependent decay of the reflectance-mode
confocal signal. The variability of these optical properties is correlated to heterogeneity of the phantom.
These microscopic properties were compared with macroscopic properties determined by ballistic
transmission experiment. This enabled to optimize the phantom preparation procedure.
Phase function is to determine the photon propagation direction change for each scattering step in Monte Carlo simulation. Henyey Greenstein function is widely used in the Monte Carlo program. In this study, we implement 3 different phase functions: Henyey Greenstein, Double Henyey Greenstein and Mie-Theory-Generated phase function. The results show that when light source is collimated beam or focused beam, there is remarkable difference of reflectance flux in the central region of incident light.