Frequency domain artifact removal technique for multistatic microwave head imaging

Microwave techniques have the potential to be used for head imaging to detect brain abnormalities. However, the outer section of the human head, which consists of different tissue layers, such as skin, skull, fat and muscles, presents a strong signal reflections and scattering clutter that might mask the target´s response. In addition, those reflections tend to persist over a long duration of time, due to the multilayer structure of the head. These artifacts debilitate microwave imaging algorithms to successfully detect brain injuries inside the head. In this paper, a hybrid artifact removal method which combines the entropy-based filtering and differential approach, is presented to remove those artifacts in multistatic microwave head imaging. In this method, an entropy-based filter separates strong reflections from other signals, whereas the differential approach removes the artifacts from those reflections at different frequency samples. The proposed approach is successfully tested using realistic multilayer head model and an 8-element antenna array.realistic multilayer head model.