Direct reconstruction of kinetic parameter images in fluorescence optical diffusion tomography

We present a technique for reconstructing the spatially dependent dynamics of a fluorescent contrast agent in turbid media. The dynamic behavior is described by linear and nonlinear parameters of a compartmental model. The method extends our previous work in fluorescence optical diffusion tomography by parametrically reconstructing the time dependent fluorescent yield. The reconstruction uses a Bayesian framework and an efficient, nonlinear, single-site update scheme closely related to Gauss-Seidel methods, which we call parametric ICD or PICD. We demonstrate the method with a simulation study, and show that it can distinguish between objects with different uptake rates with high contrast.