Abstract

Electrical Impedance Tomography is a low-cost, portable, non-invasive imaging modality that can be used to recover the internal conductivity of a body using harmless electrical measurements taken at the surface via electrodes. The mathematical problem of recovering the conductivity involves solving a severely ill-posed nonlinear inverse problem which requires carefully designed robust numerical algorithms. CGO -based methods have been shown to be robust for time-difference and absolute EIT imaging in 2D with clinical applications in real-time.   The methods require a nonlinear scattering transform tailor-made for the EIT problem which allows the problem to be solved directly without costly repeated FEM solutions.  In 3D, CGO -based EIT reconstruction has recently been achieved in a few seconds rather than the hours required by iterative optimization-based schemes.  This time savings is especially important in applications such as stroke classification and monitoring.  In this talk we explore how CGO -based reconstruction can be applied to 3D EIT demonstrating the effectiveness on experimental tank laboratory data for absolute and time-difference EIT .  Extensions to partial boundary data and natural pairings with deep learning are discussed.