Magnetic Resonance Electrical Impedance Tomography (MREIT) is an imaging modality used to obtain high resolution electrical conductivity images. In MREIT, images of electrical conductivity distribution are reconstructed utilizing the magnetic flux density distribution induced by externally applied current to a conductor media. Applying current to a conductor object during Magnetic Resonance (MR) image acquisition generates magnetic field as a function of conductivity distribution inside the object. This magnetic field acts like a local gradient field. The local gradient accumulates an additional phase in the MR signal, and by measuring the phase shift, magnetic flux density distribution can be reconstructed. Wheaton et al. recently proposed a method to extract a map of the eddy current fields from phase images obtained using a pre-stimulating pulse and spatial modulation of the magnetization (SPAMM).¹ In this study, we propose a new method for injecting current in MREIT, by first applying current in synchrony with tagging gradient of SPAMM, and then using any pulse sequence to acquire Magnetic Resonance Imaging (MRI) data. This will allow theoretically the usage of any MRI pulse sequence for MREIT data acquisition. Therefore, it can help to reduce the scan time.A phantom experiment is done on 3T Siemens Magnetom Trio System at National Magnetic Resonance Research Center (UMRAM). An 8 cm x 8 cm x 8 cm cubic phantom with 2 cm x 2 cm x 2 cm recessed electrodes on four sides. The phantom is filled with a saline solution of 0.5 S/m conductivity. A 2 cm x 2 cm x 2 cm cubic object prepared by agarose gel with 1 S/m conductivity placed at the center of the phantom, as an inhomogeneity. The experimental phantom is shown in Figure 1. Data is acquired by applying current between two orthogonal pairs of electrodes, in order to obtain a unique solution.2 20 mA current pulse is applied for 17 msec between 90° RF pulses during tagging period and signal is acquired with a following MR pulse sequence. The proposed pulse sequence during this study is illustrated in Figure 2.³ Scan parameters of MRI are given in Table 1. Conductivity distribution inside the object is reconstructed by Sensitivity Matrix Method (SMM)⁴ with Tikhonov regularization and L-curve as regularization parameter selection approach.

Magnetic flux density of vertical and horizontal profiles obtained by using method described in ¹ and the reconstructed conductivity maps are shown in Figure 3 and Figure 4b. To compare the obtained results, a numerical model is prepared and solved using Finite Element Method (FEM). The reconstructed conductivity distribution of the simulation model using SMM is given in Figure 4a. To evaluate the accuracy of the reconstructed conductivity of the measured and the simulation results an error formula is defined as follows, $$\epsilon=\sqrt{\frac{1}{N}\sum_{j=1}^N\frac{(\sigma_{true}^2-\sigma_{jr}^2)^2}{\sigma_{true}^2}}$$

where, $$$\sigma_{true}$$$ and $$$\sigma_{jr}$$$ are the true and the reconstructed conductivity distributions of the j^th pixel. N is the total number of pixels in the conductivity image. Error values of the reconstructed conductivity distribution from the numerical model and the experimental data are respectively $$$\epsilon_{num}=8$$$ % and $$$\epsilon_{exp}=19$$$ % . It is seen that the reconstructed image using MRI-SPAMM method has acceptable results in the sense of perceptual conception and the error value.

In this study, the SPAMM based MREIT data acquisition sequence is studied. Experimental results show that SPAMM current injection technique to obtain magnetic flux density data in order to reconstruct conductivity images is a viable method. By comparing the results which are obtained from the simulation model and the experimental data, it can be said that this method can provide conductivity images with a spatial resolution and perceptual conception of the MREIT method. Phantom experiments with different pulse sequences and lower injected current magnitude to further investigate the usefulness of the proposed method in reducing the scan time while increasing the SNR is underway.