Electrical Impedance Tomography

Abstract

Completely or partially disconnected electrodes are a fairly common occurrence in many EIT clinical applications. Several factors can contribute to electrode disconnection: patient movement, perspiration, manipulations by clinical staff and defective electrode leads or electronics. By corrupting several measurements, faulty electrodes introduce significant image artifacts. In order to properly manage faulty electrodes, it is necessary to 1) account for invalid data in image reconstruction algorithms and 2) automatically detect faulty electrodes. This paper presents a two-part approach for real-time management of faulty electrodes based on the principle of voltage-current reciprocity. The first part allows accounting for faulty electrodes in EIT image reconstruction without a priori knowledge of which electrodes are at fault. The method properly weights each measurement according to its compliance with the principle of voltage-current reciprocity. Results show that the algorithm is able to automatically determine the valid portion of the data and use it to calculate high quality images. The second part of the approach allows automatic realtime detection of at least one faulty electrode with 100% sensitivity and two faulty electrodes with 80% sensitivity enabling the clinical staff to fix the problem as soon as possible to minimize data loss.

A Localization Method for Untethered Small-Scale Robots Using Electrical Impedance Tomography

“…Tomography is a technique used to estimate the internal structure of an object from signals emitted and measured outside the analyzed domain [36]. Among various tomographic methods based on different physics, EIT relies on electricity to non-invasively reconstruct the interior of the target.…”

“…The electrical measurements and excitations (i.e., voltages or currents) are typically performed by means of multiple electrodes placed on the surface of the system to be imaged. This set of signals is then used to determine the distribution of the absolute conductivity (absolute EIT) within a 2D or 3D domain, or the distribution of the conductivity changes with respect to a reference state (difference EIT) [36]. This section presents the theoretical fundamentals of EIT.…”

“…where x is a point in Ω, ω is the angular frequency of the applied current, and γ is the electric admittivity given by γ(x, ω) = σ(x, ω) + iωε(x, ω), where σ is the electric conductivity and ε is the permittivity. In the conventional frequency range at which EIT systems operate, the imaginary part of the admittivity is usually negligible [36]. Therefore, γ is approximated to the conductivity σ.…”

“…Solving the inverse problem requires finding a value of the conductivity σ that minimizes the data mismatch between the electrical measurements y and their estimates via the forward solution F (σ) [36] y − F (σ)…”

“…The most popular is based on the linearization and the regularization of the problem to get an approximated well-posed problem. The equation to minimize becomes [36] y − F (σ)…”