A wideband model to evaluate the dielectric properties of biological tissues from magnetic resonance acquisitions

Objective. Aim of this work is to illustrate and experimentally validate a model to evaluate the dielectric properties of biological tissues on a wide frequency band using the magnetic resonance imaging (MRI) technique. Approach. The dielectric behaviour of biological tissues depends on frequency, according to the so-called relaxation mechanisms. The adopted model derives the dielectric properties of biological tissues in the frequency range 10 MHz-20 GHz considering the presence of two relaxation mechanisms whose parameters are determined from quantities derived from MRI acquisitions. In particular, the MRI derived quantities are the water content and the dielectric properties of the tissue under study at the frequency of the MR scanner. Main results. The model was first theoretically validated on muscle and fat using literature data in the frequency range 10 MHz-20 GHz. Results showed capabilities of reconstructing dielectric properties with errors within 16%. Then the model was applied to ex vivo muscle and liver tissues, comparing the MRI-derived properties with data measured by the open probe technique in the frequency range 10 MHz-3 GHz, showing promising results. Significance. The use of medical techniques based on the application of electromagnetic fields (EMFs) is significantly increasing. To provide safe and effective treatments, it is necessary to know how human tissues react to the applied EMF. Since this information is embedded in the dielectric properties of biological tissues, an accurate and precise dielectric characterization is needed. Biological tissues are heterogenous, and their characteristics depend on several factors. Consequently, it is necessary to characterize dielectric properties in vivo for each specific patient. While this aim cannot be reached with traditional measurement techniques, through the adopted model these properties can be reconstructed in vivo on a wide frequency band from non-invasive MRI acquisitions.