Wearable Microwave Imaging System for Brain Stroke Imaging

Abstract

This paper presents the experimental validation of the detection capabilities of a low complexity wearable system designed for the imaging-based detection of brain stroke. The system approaches the electromagnetic inverse problem via a 3-D imaging algorithm based on the Born approximation and the Truncated Singular Value Decomposition (TSVD). For testing, flexible antennas with custom-made coupling-medium are prototyped and assessed in mimicked hemorrhagic and ischemic stroke conditions. The experiment emulates the clinical scenario using a single-tissue anthropomorphic head phantom and strokes with both 20 cm3 and 60 cm3 ellipsoid targets. The imaging kernel is computed via full-wave simulation of a virtual twin model. The results demonstrate the capabilities for detecting and estimating the stroke-affected area.