Parameter estimation with discrete linear least squares method

This study investigated a parameter estimation method for the quantitation of [F-18]2-fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET) study using discrete linear least squares technique. The kinetics of FDG model is formulated as a discrete linear system. Using Z transform and its inverse transform, the rate constants of FDG model can be derived from a set of simultaneous equations in matrix form. The results show that our method is an efficient and reliable method for measuring cerebral metabolic rate of glucose. With optimal sample schedule, the error in parameter estimation is reduced and the correlation between the proposed method and nonlinear least squares method is superior to that of the original sample schedule.