Electrical Properties Tomography is a non-invasive imaging technique that allows the reconstruction of dielectric properties of human tissues, i.e. conductivity and permittivity, from the processing of magnetic resonance images. This study presents an innovative two-step approach to tackle this problem. In the first step, the goal is the simultaneous estimation of the current distribution induced by the radiofrequency (RF) source in the target and the corresponding electric field. In the second step, the constitutive relationship between the current and the field is exploited to estimate the electrical properties from the retrieved current and field distributions. Notably, the splitting of the problem into two steps allows to turn the solution of a non-linear inverse problem to the solution of a linear inverse problem, without introducing approximations. Preliminary results will be presented at the conference to demonstrate the method’s capabilities.
An Innovative Approach to MR-based Electrical Properties Tomography / Capitanio, Fabiana; Scapaticci, Rosa; Cavagnaro, Marta; Crocco., Lorenzo. - (2025). (Intervento presentato al convegno 47th Annual International Conference of the IEEE ENGINEERING MEDICINE & BIOLOGY SOCIETY tenutosi a Copenhagen, Denmark).
An Innovative Approach to MR-based Electrical Properties Tomography
Fabiana Capitanio
;Marta Cavagnaro;
2025
Abstract
Electrical Properties Tomography is a non-invasive imaging technique that allows the reconstruction of dielectric properties of human tissues, i.e. conductivity and permittivity, from the processing of magnetic resonance images. This study presents an innovative two-step approach to tackle this problem. In the first step, the goal is the simultaneous estimation of the current distribution induced by the radiofrequency (RF) source in the target and the corresponding electric field. In the second step, the constitutive relationship between the current and the field is exploited to estimate the electrical properties from the retrieved current and field distributions. Notably, the splitting of the problem into two steps allows to turn the solution of a non-linear inverse problem to the solution of a linear inverse problem, without introducing approximations. Preliminary results will be presented at the conference to demonstrate the method’s capabilities.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


