Antenna Array Optimization via Deep Learning For Breast Cancer Microwave Hyperthermia Application: Preliminary Results

Microwave hyperthermia (MH) requires the effective calibration of the antenna for selective focusing of the microwave energy at the target region with a nominal effect on the surrounding tissue. Many different antenna calibration methods such as optimization techniques and lookup tables have been proposed. In this paper, we present the preliminary results of a CNN based phase and power optimization approach. To create the necessary dataset, we used the superposition method to combine the information from the individual antennas. The results of the CNN model are compared with lookup table results. The proposed approach is promising as it shows less hot spots in heating potential distributions.