A Stroke Detection and Discrimination Framework using Broadband Microwave Scattering on Stochastic Models With Deep Learning

TL;DR

This study presents a deep learning-based framework utilizing microwave scattering data on stochastic head models for rapid, accurate stroke detection and characterization, aiming to address the urgent clinical need for portable, low-cost diagnostics.

Contribution

It introduces a novel microwave scattering approach combined with deep neural networks for simultaneous stroke detection and discrimination, bypassing traditional imaging methods.

Findings

94.60% stroke detection accuracy

AUC of 0.996 for detection

Localization error <0.004 cm

Abstract

Stroke poses an immense public health burden and remains among the primary causes of death and disability worldwide. Emergent therapy is often precluded by late or indeterminate times of onset before initial clinical presentation. Rapid, mobile, safe and low-cost stroke detection technology remains a deeply unmet clinical need. Past studies have explored the use of microwave and other small form-factor strategies for rapid stroke detection; however, widespread clinical adoption remains unrealized. Here, we investigated the use of microwave scattering perturbations from ultra-wide-band antenna arrays to learn dielectric signatures of disease. Two deep neural networks (DNNs) were used for: 1) stroke detection ("classification network"), and 2) characterization of the hemorrhage location and size ("discrimination network"). Dielectric signatures were learned on a simulated cohort of 666 hemorrhagic stroke and control subjects using 2D stochastic head models. The classification network yielded a stratified K-fold stroke detection accuracy of 94.60% with standard deviation of 2.86% and an AUC of 0.996, while the discrimination network resulted in a mean squared error of <0.004 cm and <0.02 cm, for the stroke localization and size estimation, respectively. We report a novel approach to intelligent diagnostics using microwave wide-band scattering information thus circumventing conventional image-formation.