Tr Adjoint Imaging Method for Mitat

Most of the current imaging methods in microwave induced thermoacoustic tomography (MITAT) system assume that the heterogeneous sound velocity (SV) and density distribution are given or subject to Gaussian distribution. These situations generally are not satisfied. To improve multi-targets thermoacoustic sources imaging quality in a heterogeneous tissue, an iterative TR adjoint imaging method is proposed. The proposed iterative TR adjoint method can reconstruct thermoacoustic sources from the measured data even if the prior heterogeneous information of the tissue is unknown. This method estimates misfit between synthesized and observed measured signals, and iteratively updates supposed model parameters which give the heterogeneous tissue structure. In this iterative procedure, error kernels of SV, density and the approximate point source position information can be obtained independently. After the time of flight (TOF) convergence criterion is reached, a regular time reversal (TR) method with updated model will give out the final imaging result. The proposed TR adjoint imaging method is based on strictly theoretical derivation, and some simulations are presented to validate the method. Received 24 July 2012, Accepted 9 November 2012, Scheduled 15 November 2012 * Corresponding author: Guoping Chen ([email protected]).