Study On Image Reconstruction Algorithm For Electrical Impedance Tomography

Electrical impedance imaging(EIT)is an emerging non-destructive medical diagnostic technology developed in recent decades.It has always been the focus of biomedical research and has received wide attention from all walks of life.It has the advantages of no damage,simple system,convenient operation,etc.It has far-reaching significance for the diagnosis and monitoring of the disease.This paper mainly studies the algorithm of EIT.The main work of algorithm research is as follows:Firstly,the theoretical physics model of EIT is theoretically deduced,and the finite element method is used to solve the positive problem.Secondly,the static imaging method of EIT is studied.In order to reduce the time of image reconstruction,a local fine-section model based on central region and foreign object-based region is constructed and simulated.Due to the non-uniform limitation of the local fine sectioning,the effect of image reconstruction and the anti-noise performance are degraded.In order to solve this problem,this paper degenerates the unit into nodes,uses the Newton-Raphson algorithm based on the node method to perform imaging,and reconstructs the simulated data and measured data.Experiments show that this method can improve the resolution of the image while reducing the calculation;Finally,due to the different impedance spectrum characteristics of biological tissues at different excitation frequencies,the frequency difference algorithm is also studied in this paper.Based on the traditional weighted frequency difference algorithm,the sparse constraint of L1-norm penalty is introduced to optimize it.Aiming at the solving problem of L1 norm,the paper uses the method of splitting Bregman and Gauss-Seidel iterative method to solve the problem.When the impedance of the background region varies with the excitation frequency or does not change,the experimental imaging was performed and compared with the imaging results of the traditional weighted frequency difference algorithm.The experimental results show that the optimized algorithm can accurately locate the imaging target area,the imaging is clear,and the imaging accuracy is improved.