Application Of Kalman Filtering Algorithm In Electrical Impedance Tomography

Electrical Impedance Tomography(EIT)is non-invasive,portable,low cost,and free from damage,and can response quickly and use continuously.Therefore,EIT has good application prospects in the field of medicine and industry.As the application background of EIT tends to be complex,there are higher and higher imaging requirement constantly.To improve the imaging quality through reasonable selection of image reconstruction algorithm is the research focus of current EIT.Because the traditional EIT direct solution algorithm and iterative algorithm have weak noise reduction ability,Kalman filtering algorithm was introduced.Moreover,the inverse problem solving of EIT was converted to state estimation problem.The measured value filtering was conducted through the noise statistical properties of the algorithm,and then the grey value of image was obtained.Static and dynamic target reconstruction experiment was completed based on finite element model of EIT.The advantages of Kalman filtering algorithm in terms of noise reduction were verified.With the increase of the number of measurement,the image reconstruction effect of Kalman became poorer and even image distortion appeared.The physical fields of EIT were described through nonlinear equation of Extended Kalman filtering,and then the accuracy of EIT mathematical model was improved.Next,fading factors were introduced,the weights of new measurement data in the process of solving were improved.the good numerical stability of algorithm is be guaranteed,and the quality of reconstruction image were improved.Based on experimental platform of EIDORS3.8 and baby chest CT sscannogram,priori information of baby chest outline was extracted,and three-dimensional model of baby lung was built.The lungs clinical monitoring data when baby spontaneously breathed were introduced.The lung contour image at end inspiration was reconstructed.In addition,pulmonary ventilation data were extracted and analyzed.The feasibility of Kalman filter algorithm and its extended algorithm in practical application was verified.