Effect of Excitation-detection Strategies of MIT on Frequency-differential Image Reconstruction
Zhili Xiao, Chao Tan, Feng Dong
Tianjin Key Laboratory of Process Measurement and Control
School of Electrical Engineering and Automation, Tianjin University, Tianjin, China email@example.com
In Magnetic Induction Tomography (MIT), frequency-differential image reconstruction is a promising brain imaging method that overcomes the limitation of a before-lesion data set in time-differential. To investigate the influence factor on the frequency-differential imaging, four kinds of excitation-detection strategies were numerically simulated to study the frequency-differential formulation: current excitation and amplitude detection strategy, current excitation and phase detection strategy, voltage excitation and amplitude detection strategy, and voltage excitation and phase detection strategy. 2-D finite element simulations were carried out for the four strategies, and images were reconstructed by Thikhonov regularization algorithm. The results showed that the scale parameters of frequency- differential formulations varied with the excitation-detection strategies. The scale parameter was (f1 /f2 )2 with current excitation and amplitude detection strategy, andf1 /f2 for other three kinds of excitation-detection strategies. Images reconstructed with current excitation were similar to that with voltage excitation, while the images reconstructed by phase detection were better than those by amplitude detection with either current excitation or voltage excitation. The present work implied that different excitation-detection strategies directly affect the scale parameter and then the reconstructed images, thus suggesting that the frequency-differential formulation should be chosen properly corresponding to the excitation-detection strategy in MIT reconstruction.
Keywords Excitation-detection strategies, Frequency-differential formulation, Image reconstruction, Magnetic induction tomography
Copyright © International Society for Industrial Process Tomography, 2016. All rights reserved.