Operational Learning-based Boundary Estimation in Electromagnetic Medical Imaging

TL;DR

This paper introduces a learning-based method to estimate the boundary of an object in electromagnetic medical imaging using the same data, improving efficiency and accuracy without extra sensors.

Contribution

A novel learning-based approach for boundary estimation in electromagnetic imaging that utilizes reflection coefficients, verified through clinical trials with a head imaging system.

Findings

Achieved average dissimilarity of 0.012 in Hu-moment for head boundary detection

Enabled fast scanning and imaging without additional boundary estimation devices

Validated with clinical trials using a 16-element antenna array across 0.7-1.6 GHz

Abstract

Incorporating boundaries of the imaging object as a priori information to imaging algorithms can significantly improve the performance of electromagnetic medical imaging systems. To avoid overly complicating the system by using different sensors and the adverse effect of the subject's movement, a learning-based method is proposed to estimate the boundary (external contour) of the imaged object using the same electromagnetic imaging data. While imaging techniques may discard the reflection coefficients for being dominant and uninformative for imaging, these parameters are made use of for boundary detection. The learned model is verified through independent clinical human trials by using a head imaging system with a 16-element antenna array that works across the band 0.7-1.6 GHz. The evaluation demonstrated that the model achieves average dissimilarity of 0.012 in Hu-moment while detecting head boundary. The model enables fast scan and image creation while eliminating the need for additional devices for accurate boundary estimation.