Abstract
Single-coil magnetic induction tomography (MIT) has thus far relied upon traditional bridge techniques to measure inductive losses, falling well under ~1 ohm. These measurements have been plagued by both noise and especially drift. This work considers methods based upon the Texas Instruments LDC-1101 chip, which measures LC circuit admittance while in resonance, from which we compute loss. Inductive loss is measured in a 4.0 cm diameter circular loop coil and compared with a quantitative theoretical result while the coil is axially positioned above a 2% aqueous potassium chloride solution contained in a 14 cm diameter petri dish filled to a depth of 2.0 cm. Results accurately capture the tail behavior of inductive loss as a function of coil-target distance. Then, using IR camera technology to track coil position, several `manual' scans are performed over phantoms prepared from sodium chloride-doped agarose components. In particular, this work considers the ability of single coil scans to capture corners of square plugs, gaps between plugs and side-by-side plugs that differ in conductivity. With drift and noise greatly reduced, the role of sample size is tested, showing that going beyond about 350 samples produces little further benefit to image quality. Though coil position is tracked to within ±0.25 mm, the random nature of manual positioning suggests that a more deliberate positioning scheme is needed, e.g. robotically.
Highlights
M AGNETIC induction tomography (MIT) is a modality intended to image the electrical conductivity distribution of a conductive target [1], [2]
Either of these capacitances can display unwanted losses, which can spoil the accuracy of measuring Re, a topic of the section – note that only an inductive loss originating from eddy currents is sought in our MIT application
After discretizing convolution integral (5) using deformed prismatic finite elements (1,464 elements, 783 nodes), a system of equations is produced that predicts coil loss, Aσ, at each coil position, while δ Zm is the actual set of measurements
Summary
M AGNETIC induction tomography (MIT) is a modality intended to image the electrical conductivity distribution of a conductive target [1], [2]. As with multi-coil MIT, the quality of an image produced by the reconstruction step depends very much on the distribution and number of samples obtained from a single-coil scan procedure. In previous work, these samples have been acquired. Given that newer instrumentation based upon the LDC-1101 is able to reduce drift and noise compared with older instrumentation, as this work shows, the quality of single-coil MIT results can be evaluated when the number of available samples becomes essentially unlimited. A goal of this work is to know more clearly whether lattice-like positioning of the coil is needed or if the inevitable random and arbitrary sampling associated with the IR tracking camera is acceptable. The error ε committed by considering R2 as infinite is:
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
