Supershape augmented reconstruction method for electrical impedance tomography

摘要

This paper proposes a shape reconstruction framework for electrical impedance tomography (EIT) based on the concept of supershape and Boolean operation. In the proposed framework, the basic shape primitives are estimated through a boundary description using a supershape formula. The super-shape formula allows representing a variety of shapes including shapes with smooth boundaries and/or sharp properties. In addition, to obtain topology optimization, the so-called Boolean operations are applied to the reconstruction, such that the basic shape primitives can be overlapped, merged, and separated to form various complex shapes. The performance of the proposed approach is tested with experimental water tank data and simulated data considering EIT lung imaging. Both the simulation and experimental results show that the proposed framework is capable of preserving features of inclusions and providing estimates to both the interface locations and unknown conductivity values with reliable accuracy.