The Application Of Monte Carlo Simulation In Diffuse Optical Tomography

Light propagation in biological tissue is an important research topic in the field of Tissue Optics. The research of light propagation rules in turbid media can provide a valid technical method and reliable theoretical dependence for diseases’ nondestructive diagnose and treatment on the base of optical devices.Monte Carlo simulation is a kind of numerical computation method based on the Discrete Statistical principle, which is very suitable for the area of light propagation in turbid media. It’s very attractive for its simplicity, generality and flexibility, but it gets a fatal deficiency:it’s too slow. Take these into consideration, this paper employs the Monte Carlo method to simulate the photon migration in turbid media and introduces a few methods to accelerate it.Chapter one firstly talks about the meaning of research in light propagation rules in biological tissues and establish the topic basis of this paper. This chapter also introduces the basic principles and classifications of the Diffuse Optical Tomography (DOT), as an important tool in this area, and simply describes the research history of the forward and inverse problem models.Chapter two introduces three kinds of forward problem models, including the Monte Carlo model, the Radiance Transport model and the Diffusion Equation model, and compares the advantages and disadvantages of these models.Chapter three introduces several methods to accelerate the Monte Carlo simulation. Monte Carlo simulation gets a lot of advantages as the forward problem model in DOT, but it’s too time-consuming which makes it useless. So it’s of great importance to accelerate the Monte Carlo simulation for its practical use in clinical diagnose and treatment. This chapter introduces the GPU-based Monte Carlo method to accelerate it in hardware and the white Monte Carlo (WMC) and perturbation Monte Carlo (PMC) method to accelerate it in algorithm. Then it proposes a new method named accelerated Monte Carlo method as a combination of the WMC and PMC to further accelerate the simulation. Taking the different Monte Carlo methods into practical uses, we simulate the photon migration in semi-infinite media using finger as the sample and finally succeed in getting the scattering spectra and the absorption spectra of finger in Chapter four. Based on the time costed by the different Monte Carlo methods, we compare the computation efficiency of them.Chapter five summarizes the main work of this paper and points out the work can be further carried out in future.