Simulation Research Of Arm Cross-sectional Tissue Imaging Based On EIT Technology

The principle of Electrical Impedance Tomography(EIT)is working by injecting current into the electrodes placed on the surface of the object,and measuring the external voltage response.Finally reconstructing a distribution image of electrical characteristic parameters inside the object.The advantages of its functional imaging,instruments manufactured easy,non-invasive,as well as harmless to human body.Make it have great application prospects in the fields of medical imaging.Based on the model of mathematical physics of EIT,this thesis research on the solution of forward problem and inverse problem of EIT.And the imaging test is carried out on the arm section model.The specific research contents of this thesis are as follows:1)Analysis on forward problem of EIT.According to the potential controlling equation satisfied by EIT.Using Finite Element Method(FEM)to solve the potential distribution,and a simplified upper arm model was established to test the forward problem.The calculative program of FEM is written in Matlab,and the calculation results by FEM are compared with electromagnetic field software COMSOL.The results show that the calculation results by FEM are nearly in consistent with the simulation results in COMSOL,which indicates that the accuracy and precision of FEM can be used as the basis of calculation of inverse problem.Finally,the merits and inferior between the adjacent excitation mode and the relative excitation mode,as well as the influence of finite element mesh density on the calculation precision are analyzed.2)Research on NR imaging algorithm of Modified Tikhonov Regularization.NR algorithm is applied to solve the inverse problem of EIT in this thesis.In order to solve the ill-poseness of inverse problem,the standard Tikhonov regularization method is adopted to make the solution process stable.In allusion to the problem of inaccurate size and location of target in the EIT,the regularization matrix of standard Tikhonov regularization is modified.By reconstructing images test on the arm section,the modified regularization is more accurate than the standard Tikhonov regularization on the size and location of the target,but existing insufficient distinction of target boundary as well.3)Research on Modified Tikhonov Hybrid Regularization Imaging Algorithm.In allusion to the problem that the Tikhonov regularization has insufficient target boundary distinction,the modified Tikhonov regularization and the variation regularization are combined to form a hybrid regularization imaging algorithm.The quality of the reconstructing images are greatly improved.The size and location of the target,as well as the boundary distinction are promoted correspondingly.4)Adding noise into the voltage measurement data to reconstruct images of the arm,in which the SNR is decreased to 40 d B.The reconstructed image of the arm is distorted seriously,which indicating EIT has a high requirement on the accuracy of voltage measurement data and is sensitive to the noise.