Application of SQP algorithm for fluorescence tomography with the time-domain equation of radiative transfer

A reconstruction scheme for the fluorescence tomography is investigated based on the time-domain radiative transfer equation (TD-RTE). Two coupled TD-RTEs, which can provide considerable measurement data, are used as the forward model and solved by the discrete ordinate method. The sequential quadratic programming (SQP) is employed to build the reconstruction scheme for solving the inverse problem. The gradient of objective function is calculated efficiently by the adjoint equation technique. Considering the ill-posed nature of the inverse problem, the regularization term based on the generalized Gaussian Markov random field (GGMRF) model is adopted to enhance the reconstructed image. Influence of the initial guess, contrast, noisy data, and shape of the fluorescent target are analyzed. Simulated results show that the proposed algorithm performs efficiently and accurately on reconstructing the distribution of the fluorescence yield.