Research On The Algorithm Of Human Lung Electrical Impedance Tomography

Electrical impedance tomography(EIT)is a functional imaging technique used for in vitro detection.It can extract the physiological and pathological characteristics of the human body by imaging the electrical conductivity distribution or changes of the measured object.It is compared with traditional CT and MRI,EIT has advantages of non-radiation,convenience,high imaging speed and low equipment cost.However,the current EIT system generally has problems,such as deficiencies,non-linearity,and morbidity,resulting in lower imaging quality.The change of boundary voltage is insensitive to the change of the conductivity distribution.When solving the inverse solution,if the boundary voltage changes slightly,the conductivity distribution will change greatly.Therefore,when the measurement boundary model is inaccurate and the electrode coordinates do not correspond to the measurement point model defined,it is easy to cause the EIT imaging effect.In order to solve the above problems,this paper designs a 2D telescopic boundary field measurement device for lung,which can accurately and on the basis of the 2D telescopic boundary field measurement structure,the structure and model reconstruction method for measuring the three-dimensional boundary are proposed,which are three-dimensional telescopic rod coaxial laminated measuring equipment and three-dimensional telescopic rod error stacking layered measurement equipment.Aiming at the shortcomings of fixed test points in the Electrode stack in the above two three-dimensional test structures,a three-dimensional rotatable measurement structure and a boundary model reconstruction method are proposed.The layers of this structure can revolved with each other,which can increase the visible test points on the top view plane.Interpolation algorithm improve the accuracy of the three-dimensional boundary model and provide a variety of electrode position placement methods.In this paper,the finite element method is used to solve the positive problem.Aiming at the problem that the finite element meshing will affect the accuracy of the solution,this paper increases the number of grid cells where the gradient changes greatly to improve the discernibility of the region.Aming at the periodic activity of the lungs,which leads to the inconsistency of the conductivity distribution in the lung area at different times,the difference algorithm is used to solve the conductivity change distribution before and after the time.The difference algorithm can reduce or even eliminate the noise introduced during the measurement process.Therefore this paper chooses the difference one-step Gauss-Newton method.To solve the inverse problem,which reduces the noise and imaging time,and improves the stability and accuracy of the solution of the equation system by combining the regularization technology.In order to be friendly to human-computer interaction,the operation interface is designed.The boundary measurement structure,algorithm,regularization technology and grid division method are verified by experiments.The effectiveness of boundary field measurement structure,special processing of finite element mesh and difference one-step Gauss-Newton method combined with regularization technology are verified.To improve imaging quality and imaging speed are verified.Finally,the EIT image reconstruction of the public lung measurement data is realized.