Magnetic Detection Electrical Impedance Imaging In The Study Of Pulmonary Respiratory Monitoring

There are many clinical methods used in the examination of pulmonary function,however,most of the normal methods existed are static imaging,which can not monitor the breathing process real-timely,and also can not be used to judge the information of pulmonary nidus before structural lesions.Electrical impedance tomography(EIT),based on electrical characterization,detects functional abnormalities before organic lesion,which is used for early detection of pulmonary respiratory disease.Magnetic detection electrical impedance tomography(MDEIT)is a new kind of electrical impedance tomography.It has lots of advantages,for example,non-contact,low cost,measurement convenience,and so on.In this paper,MDEIT is introduced to apply into respiratory monitoring,aiming at the forward problem of MDEIT,we set up a physical model based on lung image database consortium(LIDC)and calculate the electric potential and current density distribution inside the imaging object for normal person,emphysema patient and lung cancer patients,which using the finite element method,and we then get magnetic flux density outside the object when exhaling and inhaling according to Biot-Savart's law,then we compare them for respiratory monitoring.The results show that the magnetic induction intensity when inhaling is about 8.875%less than that when exhaling.By simulation results,we can better understand the difference of magnetic induction intensity value surrounding the lung when exhaling and inhaling due to the change of lung volume and electrical conductivity distribution.For the inverse problem of MDEIT,this paper first adopts the method of generalized inverse solution to reconstruct the distribution of current density,increasing measurement information by using the two components of the magnetic flux density.The relative errors(RE)and relative residual errors(RRE)of reconstruction of current density are numerically analyzed.On this basis,we use TSVD regularization and Tikhonov regularization to reconstruct the current density by magnetic field data with Gaussian white noise with different SNRs.The reconstruction results show that regularization can greatly improve the anti-noise performance of image reconstruction.