Microwave imaging for human brain stroke detection using frequency domain inverse modelling & phantom experiments

Abstract

Abstract This paper reports a study exploring microwave Inverse Synthetic Aperture Radar (ISAR) imaging of biological specimens, with the longer-term goal of assessing its applicability for non-invasive and non-destructive imaging of the human brain in the context of stroke detection and monitoring. The paper describes the design and fabrication of a laboratory testbed developed to examine the feasibility of the ISAR approach. The system includes a custom antenna designed to reduce self-generated clutter and support the imaging process. Water was used as a matching medium due to its specific permittivity-frequency relationship, providing controlled conditions for experimental evaluation. The forward and inverse models were initially tested in simulated environments, and subsequently evaluated using physical measurements on real biological specimens in a bistatic radar configuration, to assess their ability to localize internal anomalies with sub-centimetre resolution across a 26 cm circular imaging area. The reconstructed images from vegetable phantoms such as potatoes and turnips suggest the technique may be capable of detecting internal structural variations. These preliminary findings serve as a foundation for future investigations into human brain imaging applications.