King's College London

The use of microwave imaging for brain stroke detection has attracted growing interest in the past decade, inspired by the presence of differences in the dielectric properties of stroke and the surrounding brain tissues. This paper presents and discusses the reconstruction results from measurements on a 3D-printed anthropomorphic head model, containing a cylindrical target simulating the bleeding during a haemorrhagic stroke. To perform the measurements, the head model was immersed inside a purpose-built imaging tank containing a 90% Glycerol matching liquid, and image reconstruction has been obtained both through a DBIM-TwIST tomography algorithm and a Huygens based radar algorithm. Imaging results show that the target can be detected and localized accurately through both algorithms.