Study Of EIT Based On Improved PSO Algorithm And Improved Method Of Decomposition

Electrical Impedance Tomography is a relatively new medical imaging technology in these years. EIT technology can not only detect the changes of organs in human bodies, but also can check out the human tissues'functional changes. More importantly, contrast with other medical imaging technologies, its non-side effect when scanning human bodies is receiving more and more attentions in clinical medical areas. EIT is becoming one of the most important research topics that draw huge attentions in the interdiscipline of modern electrical techonlogy, biomedical engineering and other common fields. EIT has a wide application in future.This thesis analyses the conception, definition, basic theory and mathematical model of EIT. And this thesis puts emphasis on the forward solution and inverse solution of EIT technology.In the study of forward solution of EIT, the study in carried in such aspects including, the current injection, the number of electrodes, uneven decomposition, same area with different conductance, same conductance with different positions of areas, same conductance with different proportion of areas based on the finite element method. Base on the normally used even decomposition, this thesis improved the decomposition method, presenting a elective uneven decomposition and refined decomposition method. Choosing the unevenly decomposed area which needed to be carefully studied, the area can also be decomposed a second time. The precision of the image can be enhanced and the locating can be more locating can be more exact as well.This thesis carries out a lot of simulation experiments, including the current injection, the number of electrodes, uneven decomposition, same area with different conductance, same conductance with different positions of areas, same conductance with different proportion of areas. This thesis makes a contrast with the usual methods. Considering the actual situation, Gaussian noise and Salt & Pepper noise are added into the reconstructed images in the simulation experiments. The results turn out that basing on the improved decomposition method and particle swarm optimization, this method is able to lessen the time calculation, enhance the precision of reconstructed images in inverse solution. And the experimental results are proved to be valid and valuable in applications.