Mathematical modeling and image reconstruction in biomedical optical imaging

Diffuse optical tomography (DOT) has potential to provide clinically important functional information which is cannot measured by X-ray CT and MRI. The light is scattered by tissue and an adequate mathematical model is essential for DOT. In this project, we establish the forward model and image reconstruction algorithm based on the radiative transfer equation (RTE), which is more accurate but complicated than the conventionally used diffusion equation (DE), and a time-resolved system to measure the optical properties of tissue. A hybrid RTE-DE algorithm is developed to reduce the computational cost however keeping the high accuracy. The image reconstruction based on the RTE is evaluated by simulation and the results are clearly improved compared with that based on the DE. The feasibility study has been started to assess the applicability of RTE-based DOT to the brain and thyroid gland in which light propagation is affected by low- or non-scattering regions.