MR Imaging of Electromagnetic Field Distribution for Treatment Planning in Electrical Stimulation
Woo Chul Jeong1, Saurav ZK Sajib1, Nitish Katoch1, Bup Kyung Choi1, Hyung Joong Kim1, Oh In Kwon2, and Eung Je Woo1

1Kyung Hee University, Seoul, Korea, Republic of, 2Konkuk University, Seoul, Korea, Republic of

Synopsis

Electrical stimulations are widely used as therapeutic techniques that are closely related to the electromagnetic fields inside the human body. The electromagnetic field is affected by the injected currents and electrical conductivities of biological tissue, the map of voltage, current density, and magnetic flux density can provide meaningful information for determining the tissue type and current pathways. The signal intensity of current density is proportional to magnetic flux density which can be measured by MREIT. Since the biological tissues show anisotropic characteristic, we introduced a recent DT-MREIT method to better apply it to real situation.

Purpose

In this study, we provided the experimental results of DT-MREIT to evaluate the electromagnetic field distribution of muscle when applying the therapeutic currents into the biological tissue.

Methods

Imaging experiments were performed using chunks of bovine muscle. Two silver wire electrodes were positioned 3 cm apart inside the muscle as shown in Fig 1. Using the 3T MRI scanner (Achieva TX, Philips, Netherlands) and single-shot spin-echo EPI (SS-SE-EPI) sequence, we collected DTI data with b-values of 500 sec/mm2 and TR/TE = 4000/72 msec. One reference MR data was also obtained without diffusion sensitized gradient to measure the diffusion tensor. Using a voltage stimulator (BSLSTMB, BIOPAC Systems, Inc., USA), electric pulses were applied through the electrode pair with 100 V amplitude and 100 μs width for MREIT data collection. We used the multi-gradient echo pulse sequence to obtain the induced magnetic flux density data called Bz data. The imaging parameters were as follows: TR/TE = 200/1.6 msec, ES = 2.3 msec, NE = 32, FOV = 200*200 mm2, imaging matrix = 128*128, imaging time = 51.2 sec (Fig. 2).

For the current density estimation, we adopt the projected current density method1 which approximately recovers the best three-dimensional current density using the measured data of Bz. Based on the model about the relation between the water diffusion tensor and the conductivity tensor, we set the conductivity tensor as a scalar multiple of the diffusion tensor. Since diffusion tensor information is available from the acquired DTI images, we computed the position-dependent scale factor using the measured Bz data. Then, the conductivity tensor is estimated as the multiplication of the water diffusion tensor and the scale factor. Once we obtained both the current density as a 3*1 vector and the conductivity tensor as a 3*3 matrix, we computed the electric field by multiplying the inverse of the conductivity tensor to the current density vector.

Results and Discussion

Figure 3(a) and (b) show the acquired MR magnitude and Bz images of the bovine muscle, respectively. Figure 3(c) plots the computed projected current density image when the therapeutic currents are injected. The current density image indicates the internal current flows which exist not only the electrodes but also surrounding regions. Figure 4(a) is the recovered electric filed map using our anisotropic conductivity tensor with the projected current density whereas (b) shows the electric field map produced by using the method of Kranjc et al.2 Comparing two methods, projected current density method shows enhanced signals in both the electrodes and surrounding tissues by considering the anisotropy of muscle.

Conclusion

Electric field map obtained by using the DT-MREIT method is significantly different from the one obtained by using the previous simpler method. Since the accurate electric field mapping is important to correctly estimate the coverage of the electrical treatment, future studies should include more rigorous validations of the new method through in vivo and in situ experiments.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2013R1A2A2A04016066, 2014R1A2A1A09006320)

References

1. Kwon OI, Jeong WC et al. Anisotropic conductivity tensor imaging in MREIT using directional diffusion rate of water molecules. Phys. Med. Biol. 2014;59:2955-2974.

2. Kranjc M, Bostjan M et al. In situ monitoring of electric field distribution in mouse tumor during electroporation. Radiology 2014;274:115-123.

Figures

Fig. 1. Experimental setup for muscle tissue phantom (left) and schematic diagram of measuring the electrical stimulation (right).

Fig. 2. Single-shot spin-echo EPI sequence (left) for DTI and multi-gradient echo pulse sequence (right) to image the electromagnetic field distribution by injecting therapeutic currents.

Fig. 3. MR magnitude (a), magnetic flux density (b), and estimated projected current density (c) images of muscle phantom.

Fig. 4. Comparison of electromagnetic field distribution of muscle tissue. Anisotropic conductivity with projected current density (a) and isotropic conductivity with approximate current density (b).



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1562