Introduction

Bio-degradable Mg-based implants¹ promote patient comfort by eliminating implant removal surgery. Post implantation, the degradation kinetics of the Mg alloys might vary for different environments/patients. Therefore the healing process of the surrounding bone/tissues and the corrosion state of the implants need to be monitored². MRI may be exploited to examine the degradation status of biodegradable implants and to study the implant/tissue interface³. The metallic nature of conductive Mg implants constitutes challenges for MRI. This includes 1) MR safety risks due to the elevation of the specific absorption rate (SAR) in the vicinity of implant⁴ and 2) B₁ field inhomogeneities caused by interferences of the radio frequency (RF) field with the implant. These constraints can be offset by leveraging the degrees of freedom facilitated by parallel transmission using multichannel RF coil arrays. Recognizing this opportunity, this work examines the applicability of RF array configurations for safe MRI of biodegradable implants at 7.0T. For this purpose RF array configurations comprising loop elements and/or fractionated-dipoles⁵ are characterized in electromagnetic field simulations using the metrics peak-SAR_10g and transmission field uniformity to derive optimum B₁ shim patterns using genetic algorithm (GA). The results are benchmarked against the birdcage mode⁶ (CP) and the orthogonal projection method⁷ (OP) in simulations and in a phantom study with a focus on B₁⁺, specific absorption rate (SAR) and temperature maps.

Methods

EMF simulations were performed (CST Studio Suite 2020, CST MWS, Darmstadt, Germany) for three cylindrical array configurations of: (i) 8 loop elements, (ii) 8 dipoles, (iii) 8 building blocks comprising a loop and a dipole element⁸. The evaluation focused on achievable implant SAR reduction and homogenization of B₁⁺ in a cylindrical (r=20mm and L=110mm) region of interest (ROI) around the implant. EMF shimming was optimized using a multi-objective genetic algorithm that calculates the tradeoff-curve of B₁ homogeneity in the ROI vs. the implant SAR. The outcome of the GA was benchmarked against CP and OP shimming patterns. In simulations, a cylindrical phantom (r=85mm and L=300mm) mimicking muscle electrical properties at 297.2 MHz was used. The same PVP-based phantom⁹ was used for the experiments A 70mm long copper wire with 1mm radius was used to mimic an implant and aligned in parallel with the phantom at a 30mm distance from its surface. The RF modules were positioned with 20mm distance from the phantom and shielded with 30mm distance from the elements. Phantom experiments were performed on a 7T MRI scanner (SIEMENS Magnetom) where the RF transceiver array was connected to the scanner through an interface (MRI.TOOLS GmbH, Germany) to utilize the pTx mode. B₁⁺ maps were acquired using a fast gradient-echo imaging with pre-saturation¹⁰. Characterization of implant heating was conducted with SAR mapping (simulations) and temperature mapping (experiments) using a high-power heating regime (P_avearage=175W, t=5 min).

Result

Our EMF simulations revealed that an array of 8 RF building blocks comprising a loop and dipole element as shown in figure 1 is the superior design. We obtained excellent agreement between the measured and simulated B₁⁺ maps as outlined in figure 2 and 3, respectively. The GA approach outperforms the CP and OP in the ROI in terms of both B₁⁺ magnitude and homogeneity as well as SAR efficiency. Statistical parameters of the simulated B₁⁺ in the ROI are presented in the table of figure2. The SAR maps of CP, OP and GA maps are presented in figure 4 and the CP and GA temperature maps in figure 5. The simulated value of maximum average SAR_10g (normalized to 1W excitation) in the whole phantom using CP, OP and GA is 0.2 $$$ {1}/{kg}$$$ , 0.31 $$$ {1}/{kg}$$$ and 0.24 $$$ {1}/{kg}$$$, respectively. For the CP mode, the maximum is found at the implant, whereas for both OP and GA excitation, peak SAR occurs at the surface. Even though overall peak SAR for unit power is slightly increased comparing GA to CP excitation, B₁⁺ in the ROI is also significantly increased, leading to $$${B^+_1}/{\sqrt {SAR_{Max}}}$$$ efficiencies of 0.54 $$${\mu T}/{\sqrt {{W}/{kg}}}$$$ and 0.67 $$${\mu T}/{\sqrt {{W}/{kg}}}$$$ for CP and GA, respectively, i.e. showing a 24% efficiency increase. While OP can significantly reduce implant SAR, it nevertheless suffers from a peak SAR increase of more than 50%. On the other hand, the GA method successfully limits both implant as well as overall peak SAR while still delivering improved B₁ in the ROI. The maximum temperature increment in the experiment using GA is 61% lower compared to CP.

Discussion

CP excitation leads to B₁ artifacts close to the implant as well as excessive local RF power deposition. The orthogonal projection method reduces implant heating but suffers from B₁ degradation in the implant vicinity as well as excessive peak SAR increase outside the implant region. Our results demonstrate that the multi-objective GA approach affords SAR reduction in conjunction with a strong and uniform B₁⁺ field in the implant region.

Conclusion

Parallel transmission using the proposed 8-channel RF array in conjunction with the GA approach helps to meet the requirements of MR safety and transmission field uniformity to support MRI based monitoring of tissue healing and for the benefit of MRI characterization of the degradation state of bio-degradable implants.

Acknowledgements

This project is supported by the H. 2020 European Training Network of the European Union (MgSafe, grant agr. No. 811226)

References

[1] B. J. C. Luthringer, F. Feyerabend, and R. W. Römer, “Magnesium-Based Implants: A Mini-Review,” Magnesium Research. 2014, doi: 10.1684/mrh.2015.0375. [2] A. Myrissa et al., “In vitro and in vivo comparison of binary Mg alloys and pure Mg,” Materials Science and Engineering C, 2016, doi: 10.1016/j.msec.2015.12.064. [3] D. Paramitha, M. F. Ulum, A. Purnama, D. H. B. Wicaksono, D. Noviana, and H. Hermawan, “Monitoring degradation products and metal ions in vivo,” in Monitoring and Evaluation of Biomaterials and their Performance In Vivo, 2017. [4] L. Winter, F. Seifert, L. Zilberti, and M. Murbach, “MRI-Related Heating of Implants and Devices : A Review Physics of Implant Heating,” pp. 1–20, 2020, doi: 10.1002/jmri.27194. [5] A. J. E. Raaijmakers et al., “The fractionated dipole antenna: A new antenna for body imaging at 7 Tesla,” Magnetic Resonance in Medicine, 2016, doi: 10.1002/mrm.25596. [6] V. Alagappan et al., “Degenerate mode band-pass birdcage coil for accelerated parallel excitation,” Magnetic Resonance in Medicine, 2007, doi: 10.1002/mrm.21247. [7] L. Winter et al., “Parallel transmission medical implant safety testbed: Real-time mitigation of RF induced tip heating using time-domain E-field sensors,” Magnetic Resonance in Medicine, 2020, doi: 10.1002/mrm.28379. [8] M. A. Ertürk, A. J. E. Raaijmakers, G. Adriany, K. Uğurbil, and G. J. Metzger, “A 16-channel combined loop-dipole transceiver array for 7 Tesla body MRI,” Magnetic Resonance in Medicine, 2017, doi: 10.1002/mrm.26153. [9] S. Chung, D. Kim, E. Breton, and L. Axel, “Rapid B1+ mapping using a preconditioning RF pulse with turboFLASH readout,” Magnetic Resonance in Medicine, 2010, doi: 10.1002/mrm.22423.