Clutter Removal Techniques for Medical Microwave Imaging

Le résumé fourni par la source

Microwave imaging has emerged as a promising modality for various biomedical applications, offering advantages such as portability, non-ionizing radiation, cost-effectiveness, and real-time scanning. However, clutter, unwanted signals from strong reflections, and different types of tissue interactions complicate imaging and hinder accurate diagnosis. This study provides a comprehensive review and comparative analysis of clutter removal algorithms in microwave imaging. Traditional clutter-removal methods, such as differential subtraction, average subtraction, symmetric subtraction, and adjacent subtraction, have been widely used for their fast processing and simplicity but often fall short in producing high-quality, clutter-free images. More sophisticated methods, such as Empirical Mode Decomposition (EMD)-based, Singular Value Decomposition (SVD)-based, spatial filtering, entropy-based, and entropy-Wiener filter-based techniques, offer improved performance but still do not meet clinical standards. To guide and motivate researchers working in this area, this review not only discusses clutter removal algorithms, but also investigates the performance of key algorithms across various environments, from simple homogeneous to complex heterogeneous domains, and highlights those used in clinical environments. This review also suggests that AI methods guided by the physics of the problem could offer a potential solution; however, they are computationally and data-intensive. This is a challenge considering the limited clinical data from microwave imaging systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.