Experimental Validation of a Microwave Scanner for Brain Stroke Monitoring in Realistic Head Models

Abstract

This work addresses brain stroke evolution assessment in mimicked clinical conditions using a low-complexity microwave imaging (MWI) scanner and realistic anthropomorphic head models. In particular, the MWI prototype employs a wearable 22-element flexible-antenna array, keeping a simple architecture and demanding low-computing power. It allows a real-time follow-up of the stroke-affected areas, providing 3-D maps of the dielectric variation through a differential linear imaging approach based on the Truncated Singular Value Decomposition (TSVD), the distorted Born approximation, and artifact removal procedure. The system includes a digital twin that emulates high-fidelity scenarios via EM full-wave simulations performed in an in-house Finite Element Method (FEM) solver. Finally, the assessment examines the system's monitoring capabilities involving custom-made and lifelike phantoms representing a dynamic stroke evolution.