Electrical Impedance Tomography using Differential Evolution integrated with a Modified Newton Raphson Algorithm

Electrical impedance tomography (EIT) is a non-invasive medical imaging procedure. Image reconstruction in EIT is difficult because it involves solving a non-linear and ill-posed mathematical problem. One of the most commonly implemented inverse approaches is usually a variation of the Newton Raphson algorithm. However, this approach is not guaranteed to reach a global optimum or a local optimum and as such, it requires an accurate initial estimation of the resistance distribution, which is not always available in practice. In this paper, a new method is proposed to solve for the inverse problem in EIT while avoiding dependencies on the initial estimation of the resistance distribution. The proposed approach uses a differential evolution (DE) optimizer integrated with the Newton Raphson algorithm. The stochastic nature of DE allows the problem to be solved without having an accurate initial estimation and allows for solutions that will not be trapped in local minimal values. Simulation results indicate that the proposed approach outperforms the traditional differential evolution algorithm, and performs similarly to the traditional Modified Newton Raphson algorithm with accurate initial estimation. The proposed method does, however, have an advantage over the Modified Newton Raphson algorithm as it does not require an estimate of the initial resistance distribution.