Frequency-difference EIT (fdEIT) using weighted difference and equivalent homogeneous admittivity: validation by simulation and tank experiment.

A new method for producing frequency-difference images in electrical impedance tomography (EIT) has been recently suggested. It employed the use of a weighted voltage difference between two frequencies. In this paper, we first explain why the weighted difference is advantageous for some applications of the frequency-difference EIT (fdEIT). Based on a relationship between injection currents at two frequencies and a weighted difference of two corresponding complex voltages, we establish an fdEIT image reconstruction algorithm. In order to apply the algorithm to a practical setting, we propose the concept of an equivalent homogeneous admittivity whose value can be estimated by measuring induced voltages at the third frequency. To test this new fdEIT algorithm, we performed numerical simulations and imaging experiments using two-dimensional phantoms with frequency-dependent admittivity distributions. From reconstructed real- and imaginary-part fdEIT images, we could validate its advantage in terms of visualizing anomalies with fewer amounts of artifacts. We propose the method for applications in tumor or stroke imaging where we are mainly interested in contrast information within an fdEIT image. We suggest investigating the forward and inverse problems of an imaging domain with a frequency-dependent admittivity distribution, which has not been addressed rigorously until now.