Effect of Varying Prior Information in Axillary 2D Microwave Tomography

• Autor(a) principal: Savazzi, Matteo
• Data de Publicação: 2022
• Outros Autores: Karadima, Olympia, Felicio, Joao M., Fernandes, Carlos A., Kosmas, Panagiotis, Conceicao, Raquel C.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10451/53367
• topic: axillary lymph node imaging; breast cancer; distorted Born iterative method (DBIM); microwave tomography; prior information

We numerically assess the potential of microwave tomography (MWT) for the detection and dielectric properties estimation of axillary lymph nodes (ALNs), and we study the robustness of our system using prior information with varying levels of accuracy. We adopt a 2-dimensional MWT system with 8 antennas (0.5-2.5 GHz) placed around the axillary region. The reconstruction algorithm implements the distorted Born iterative method. We show that: (i) when accurate prior knowledge of the axillary tissues (fat and muscle) is available, our system successfully detects an ALN; (ii) ±30% error in the prior estimation of fat and muscle dielectric properties does not affect image quality; (iii) ±7mm error in muscle position causes slight artifacts, while ± 14mm error in muscle position affects ALN detection. To the best of our knowledge, this is the first paper in the literature to study the impact of prior information accuracy on detecting an ALN using MWT.