Image Reconstruction Algorithm In MAT-MI With Acoustic Inhomogeneous Media

Magnetoacoustic Tomography with Magnetic Induction (MAT-MI) is a new functional imaging method, which take advantage of the good contrast of Electrical Impedance Tomography, together with the good spatial resolution of Ultrasonography. Owing to a multiphysics coupling technology, both of acoustic characteristics and electrical characteristics should be taken into consider in the study of MAT-MI. This paper discussed the propagation of acoustic wave under the impact of different acoustic characteristics in MAT-MI. The acoustic inhomogeneous medium, meanwhile, was considered in the reconstruction problem.Firstly, on the basis of the magnetoacoustic coupling theory and acoustic basic equation, we deduced the formulas to solve the forward and inverse problems in the condition of acoustic homogeneous and heterogeneous properties. After that, we proposed a solution method for acoustic wave equation in the inhomogeneous medium, based on finite difference time domain method. The electrical impedance model was created with COMSOL multiphysics to solve the electromagnetic forward problem. Based on the results of electromagnetic forward problem, we used Finite Difference Time Domain (FDTD) to study the effect of acoustic characteristic in MAT. Results shew that the non-homogeneous acoustic property revealed the difference in the propagation of the acoustic wave, such as the reflection, refraction and transmission phenomenon, which made the sound field problem complicated.Secondly, a scan pattern was designed by using linear array transducer. Algebraic Reconstruction Technique (ART) came out synchronously to reconstruct the acoustic speed in the acoustic heterogeneous medium. Then, FDTD was adopted to carry out the reconstruction of acoustic sources. In the simulation, two models with the same electrical impedance characteristics but different acoustic characteristics were established for evaluating the feasibility of the algorithm. Reconstruction results were agreeable with the distribution of acoustic speed and acoustic source correctly.Finally, in order to estimate the accuracy of these algorithm, we created a 3D model, whose three different sections were involved to verify the effect of the reconstruction result with more complex structures, in addition with different noise level. At the same time, a special models were established to demonstrate the ability to solve problems of electrical property as well as acoustic property As a result, the algorithm in this paper could reconstruct the distribution of acoustic speed and acoustic source correctly, it was capable of possessing either certain electrical impedance characteristics or acoustic characteristics. Moreover, we compared the proposed reconstruction method with the existing method, the proposed one was technically superior.In summary, a new algorithm was proposed for forward problem in MAT-MI with acoustic inhomogeneous medium, which laid some groundwork for studying impact of acoustic inhomogeneity. For the inverse problem, this method was firstly proposed to obtain reconstruction in acoustic inhomogeneous media without involving other imaging data. What is more, the reconstructed distribution of acoustic speed was same as that of the ultrasonic image, which could distinguish the boundaries of acoustic impedance clearly. Notable, it might give a new way to the processing of MAT-MI, for that an ultrasonic image could be obtained accompanied with the distribution of electrical impedance.