Image Reconstruction Using Particle Swarm Optimization (PSO) in Electrical Impedance Tomography

Electrical impedance tomography (EIT) aims in reconstructing resistivity distribution of inhomogeneous objects, by using the voltage measurements from the boundary electrodes and facilitates in solving non linear inverse problems. For the medical imaging research community, developing a suitable reconstruction algorithm is a challenging task as the inverse problem in EIT is severely ill-posed. This paper proposes a heuristic particle swarm optimization (PSO) method for solving the static EIT inverse problems. Experiments are performed using 64 finite elements mesh by varying the swarm size, and the reconstruction performances are evaluated in terms of the fidelity measure such as mean relative error (MRE) which indicates the error between the estimated and true values of resistivity. It is observed from the simulation results that the optimization is attained for the swarm size of 28 for which the reconstruction error is minimum. Further the spatial resolution is improved in larger extent compared to that of modified Newton – Raphson method at the outright of relatively expensive computational time. Keywordsparticle swarm optimization (PSO), electrical impedance tomography (EIT), inverse problem.