RBF neural network image reconstruction for electrical impedance tomography

Reconstruction of images in electrical impedance tomography requires the solution of a nonlinear inverse problem. This work presents a RBF neural network image reconstruction method trained by the genetic algorithm. The genetic algorithm is used to search for the optimum values of the following three parameters in the RBF network: centers, variances and connection weights, which are encoded as real number. Experimental results illustrate that this method can markedly improve image quality.