Low Computational Demand Nonlinear Correction of the Inverse Problem in Microwave Brain Imaging

This paper investigates a nonlinear correction factor for the direct inverse problem solution in medical microwave imaging (MMI), focusing on acute brain stroke monitoring. The correction factor relies on a pseudo-Rytov approximation, which employs the ratio between total and incident electric fields in the scattering model to enhance quantitative accuracy. This approach enables the direct correction of the approximate linearized imaging kernel without requiring iterative computations of the direct scattering model, significantly reducing the inversion computational effort and improving the system's robustness to numerical inaccuracies. MMI represents a promising modality for fast, potentially real-time response, delivering quantitative insights that complement gold-standard imaging techniques. This study presents a realistic numerical experiment for hemorrhagic stroke detection, demonstrating the proposed correction's impact on the accuracy of dielectric contrast reconstruction within a 3-D imaging framework and underscoring its potential benefits for clinical applications.