Microwave Tomography Bone Imaging: Analysing the Impact of Skin Thickness on the Reconstruction of Numerical Bone Phantoms

Microwave Tomography (MWT) imaging is currently being investigated as a new diagnostic modality for assessing bone health conditions. It has already been employed in other medical contexts, showing different advantages (i.e., portability, low cost, and use of non-ionizing radiation) compared to the current imaging modalities. Early investigations of MWT for bone imaging have considered the bone without the skin layer even in numerical investigations. Therefore, in this study, a realistic three-layer calcaneus-shaped bone phantom involving an external skin layer has been developed to investigate the effect of the skin layer on the reconstruction of bone images using MWT. Simulated data were obtained using a circular array of 16 antennas surrounding the phantom. To solve the electromagnetic (EM) inverse scattering problem, the Distorted Born Iterative Method (DBIM) was used in combination with an Iterative Method with Adaptive Thresholding for Compressed Sensing (IMATCS). An L 2 -regularization was adopted in the IMATCS algorithm. The goal of this study was to assess whether numerical reconstructions can be achieved by varying the thickness of the outer skin layer of the phantom. The same phantom was tested with four different skin thicknesses ranging from 1 mm to 4 mm and without the presence of a skin layer. The results suggest that it is possible to successfully reconstruct the inner bone tissues (i.e., trabecular and cortical layers) properties using the proposed method. However, accurately reconstructing the shape and dimension of the reference becomes challenging when the thickness of the skin layer exceeds 2 mm. This work highlighted the need to include the skin layer within the model in numerical investigations because its influence is worth considering to accurately replicate realistic clinical scenario where human skin is present.