Research On Temperature Distributions And Optimization Of Antenna Array Excitations In Microwave Hyperthermia For Breast Cancer

Hyperthermia treatment is the fifth tumor therapy after surgery, radiotherapy, chemotherapy and biotherapy. Clinical studies have shown that hyperthermia can not only kill cancer cells directly, but also improve therapy effects when delivered as an adjuvant to radiotherapy and/or chemotherapy due to its synergistic sensitizing effects and little side effects. In addition, microwave hyperthermia can improve organism immunity to achieve better long-term therapeutic effects of tumor treatment.With the temperature distributions directly reflecting the hyperthermia treatment effects, it has guiding significance to design the antenna arrays. To obtain more detailed temperature information,3-D (three-dimensional) three-layer model, which included skin, tumor and fat, was build in this study. Antenna model mesh generation based on its real structure and single-gap feed mode was used for antenna modeling more reasonable. Computations of electromagnetic field and temperature field are solved by self-written FORTRAN program based on FDTD (finite-difference time-domain method) and finite difference approximations of Pennes biological heat-transfer equation respectively.3-D temperature distributions were given by coupling calculations of electromagnetic field and temperature field.Genetic algorithms (GA) have advantages of population search, and do not need objective function and criterion of transfer probability. Meanwhile the solution obtained by GA is global optimal solution. In this paper, GA was used to optimize the amplitudes and phases of antenna arrays excitations for more effective hyperthermia. The corresponding temperature distributions based on the optimal excitation were given. First two of three antennas were set to the same excitations, fewer parameters can be needed to optimize if good hyperthermia effects can be obtained. However, the GA optimal results were worse than the. Then three antenna excitations to be optimized were all as independent parameters and the objective function was improved. The second GA optimal results were better than the first from the 3-D temperature distributions. It can be seen that the antenna optimal parameters selection was very important.