摘要

In order to reduce the effects of acoustic heterogeneity and therefore accurately reconstruct the microwave absorption properties, a large number of measured data are required to estimate sound velocity distribution in microwave-induced thermoacoustic tomography (MITAT). However, the process of estimating the sound velocity distribution of heterogeneous tissue will consume lots of computing time and memory. This problem becomes serious especially for three-dimensional (3-D) imaging. To solve this problem, an improved iterative reconstruction method is proposed. Based on the variation trends of time-of-flights (TOFs) of ultrasonic signals, the spatial area where the acoustic heterogeneity of tissue may exist is estimated. Then the optimal TOFs are selected according to the estimated heterogeneous area. Using the selected optimal TOFs, the sound velocity distribution of tissues is iteratively estimated until convergence occurs. Finally, the microwave absorption distribution of tissues is reconstructed with the estimated sound velocity distribution. Contrary to the traditional iterative reconstruction method, which uses all the TOFs, the proposed method only uses the optimal TOFs to estimate the sound velocity distribution of tissues. It efficiently reduces the redundant information and therefore saves computing time and memory. Both numerical simulations and data experimentally measured from MITAT system are used to demonstrate the capability and effectiveness of the proposed method.