Alternating minimization of the negative Poisson likelihood function for the global analysis of fluorescence lifetime imaging microscopy data

We investigate a fast optimization method for determining the minimizer of the negative Poisson likelihood function for the global analysis of fluorescence lifetime microscopy. Using the alternating optimization strategy, we iteratively solve a non-convex optimization problem to estimate the lifetime parameters and a convex optimization problem to estimate the concentration parameters. We effectively determine the minimizer of the non-convex optimization using the Gauss-Newton method and that of the convex optimization by applying the optimization transfer strategy, which is based on the convex inequality. In the simulation studies, the proposed method was able to determine the minimizer of the objective function significantly faster than the conventional simultaneous optimization method.