Premium
Model‐based microwave image reconstruction: simulations and experiments
Author(s) -
Ciocan Razvan,
Jiang Huabei
Publication year - 2004
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.1812871
Subject(s) - microwave imaging , microwave , iterative reconstruction , tikhonov regularization , finite element method , antenna (radio) , materials science , iterative method , computer science , acoustics , optics , physics , algorithm , inverse problem , computer vision , mathematics , mathematical analysis , telecommunications , thermodynamics
We describe an integrated microwave imaging system that can provide spatial maps of dielectric properties of heterogeneous media with tomographically collected data. The hardware system ( 800 – 1200 MHz ) was built based on a lock‐in amplifier with 16 fixed antennas. The reconstruction algorithm was implemented using a Newton iterative method with combined Marquardt‐Tikhonov regularizations. System performance was evaluated using heterogeneous media mimicking human breast tissue. Finite element method coupled with the Bayliss and Turkel radiation boundary conditions were applied to compute the electric field distribution in the heterogeneous media of interest. The results show that inclusions embedded in a 76‐diameter background medium can be quantitatively reconstructed from both simulated and experimental data. Quantitative analysis of the microwave images obtained suggests that an inclusion of 14 mm in diameter is the smallest object that can be fully characterized presently using experimental data, while objects as small as 10 mm in diameter can be quantitatively resolved with simulated data.