Study on PSO-tGN algorithm of bio-electrical impedance tomography system

Newton-like algorithm is one of usual methods to solve inverse problem of electrical impedance tomography (EIT), but it is sensitive to initial values. In order to solve the above disadvantage, a hybrid algorithm based on Particle Swarm Optimization algorithm and regularization Gauss-Newton algorithm (PSO-tGN) is proposed, in which the Particle Swarm Optimization algorithm is firstly used to get the initial electrical impedance distribution and then the regularization Gauss-Newton algorithm is used to solve the problem iteratively. The two dimension circular domain is used as the study object to realize image reconstruction of single simulation object. During the image reconstruction process, the Finite Element Method (FEM) is used to divide the measured circular domain, and the adjacent current pattern is adopted. Simulation results show that the PSO-tGN algorithm can accurately reflect the electrical impedance distribution and locate the position of target.