Enhanced Particle Swarm Optimization Algorithm for EIT Image Reconstruction

Electrical Impedance Tomography (EIT) is a non-invasive, radiation-free imaging method that has witnessed an intensive rise of interest in the biomedical field. To reconstruct an image a boundary voltages are measured using a set of electrodes, then based on these values the inner conductivity distribution is reconstructed. Nevertheless, EIT image reconstruction suffers from low spatial resolution as it represents a highly ill-conditioned, and nonlinear inverse problem. Besides, most of the existing population-based methods, used to solve this problem, get easily trapped in the local optima, which leads to an inaccurate image. This work proposes an optimized swarm intelligence algorithm based on Particle Swarm optimization (PSO) for EIT image reconstruction, as it considers the EIT inverse problem as an optimization problem and makes use of the whole search space to find the global optima. Two variations of PSO with a non-blind search have been developed and tested, namely an Advanced Particle Swarm optimization (APSO) and a Refined Advanced Particle Swarm optimization (RAPSO), to improve the performance of the PSO algorithm for EIT image reconstruction. The implementation of these algorithms is carried out on a circular mesh. After several tests, RAPSO proved a better outcome, with 0.024 fitness in 200 iterations and a population size equal to 100.