EIT Forward Computation Based On Element Free Method For Hematoma Detection

Electrical impedance tomography(EIT)is a functional imaging technology,which uses the human body impedance to carry the rich physiological and pathological information on the human cardiovascular,lung,stomach and other organs of the dynamic continuous image observation,with potential clinical application prospects.Accurately solving the EIT problem is the basis of image reconstruction.In this thesis,the EIT forward problem is calculated by element-free method(EFM)instead of finite element method(FEM),for EFM requires only the node information and no unit information.The feature of FEM is suitable for the simulation of human small lesion target.It is expected to achieve greater accuracy in the process of simulating the change in electrical impedance during hematoma detection of the human spleen.In this thesis,after analyzing the characteristics of several typical EFMs,the element-free Galerkin method(EFGM)is used to calculate the EIT forward problem.Then,the various types of weight functions are analyzed,and the simple quadrilateral spline function is chosen as the weight function,which can guarantee the accuracy without determining the variable information artificially.According to the characteristics of the influence function and the calculation time of the CPU,the value of the adjustable coefficient in the radius is determined,which ensures the accuracy and speed of the solution process.At the same time,the selection of the impose boundary conditions and the numerical integration scheme are all as much as possible to improve the accuracy of the calculation under the positive definite equation.On the basis of theoretical analysis,the model of human spleen hematoma is established,and the simulation results are obtained.The results are compared with that calculated by FEM.Finally,traditional back projection algorithm(BPA)and node back projection algorithm(NBPA)are used to solve the inverse problem.Through the quantitative evaluation parameters,it is found that the results obtained by EFGM are smoother,more complete than those calculated by FEM.And the image calculated by EFGM contains more information.