4187

Radiofrequency-Induced Risks during Magnetic Resonance Imaging: Dependence upon Birdcage Design

Aiping Yao¹, Manuel Murbach², Tolga Goren¹, and Niels Kuster^1,3
¹IT'IS Foundation, Zurich, Switzerland, ²UPV, Valencia, Spain, ³ETH Zurich, Zurich, Switzerland

Synopsis

The induced radiofrequency (RF) electrical field resulting in unwanted heating in patients with and without implants during magnetic resonance imaging (MRI) is a complex function of multiple factors which must be considered in patient risk assessment, including patient anatomy, imaging positions, RF characteristics of the implant, implant location, and MRI scanner design and scan configuration. In this paper, the dependence of the induced fields upon the birdcage design is investigated to determine the minimal range of RF body coils necessary to guarantee a comprehensive risk evaluation regarding heating and electromagnetic compatibility for any clinical MRI scanner.

Introduction

For patients undergoing magnetic resonance imaging (MRI), RF-induced heating is one of the major safety concerns, particularly for patients with long, conductive implants¹. The localized RF-induced field distributions can only be reliably estimated by simulation^1-4 with computational high-resolution anatomical human body phantoms, such as the Virtual Population⁵ placed at different landmark positions in a birdcage models. Risk assessment of patient heating requires that the range of possible induced fields experienced by patients in clinical scanners be considered.

In this work, we performed a sensitivity analysis of the role of birdcage geometry and structure on the induced fields and specific absorption rate (SAR) as well as tangential E-fields (E_tan) to, and deposited power (P_dep) around, representative generic active medical implants (pacemaker, and spinal cord stimulator). We then compared the results to the variation observed from patient anatomy, scan position, and scan polarization, at 64 MHz.

Method

Three studies are designed. Study 1 uses the MRIxViP1.5T database⁶ to extract the head SAR, whole-body SAR, partial-body SAR, and B₁₊ for different anatomical phantoms as well as the E_tan to, and P_dep around, a 40-cm long generic left-side pacemaker, for a set of ten birdcages, eight anatomical models, all imaging positions from head to foot, at circular and near-circular polarizations, as shown in Figure 1 (a). The role of birdcage diameter and length are assessed based on these results.
In study 2, a set of additional birdcages were developed to estimate the sensitivity of these results to different design features, namely high-pass and band-pass birdcages fed from legs and via I/Q channels (Figure 1(b)). The patient position sensitivity was estimated by shifting model Fats 2.5 cm laterally or 10 cm dorsally.

The goal of study 3 was to investigate the effect of the number of birdcage rungs on the in vivo RF-induced heating. Birdcages with length of 70 cm and bore-diameter of 70 cm with rung numbers (N) of 8, 16, and 32 were used to simulate the exposure of human model Duke at a typical thorax imaging position. The magnetic field strength and the induced in vivo electrical field distributions at the imaging position's field-of-view (FOV) are compared.

Results

Figure 1 summarizes the exposure configurations of the birdcages and anatomical models used in this work.
First, the comparison of head SAR (hSAR), whole-body SAR (wbSAR), and partial-body SAR (pbSAR) under normal operating mode⁷ are reported together with the maximum allowed magnetic field (B₁) exposure level. Figures 2 and 3 summarize the results of study 1 for birdcage length and bore diameter sensitivity, respectively. As shown in Figure 2, the longer birdcage lengths result in larger SARs under normal operating mode, leading to lower B₁ field strength at the SAR limit. In contrast, Figure 3 shows that the effect of birdcage bore-size on the in vivo SAR distribution is very limited, as birdcages with different bore-diameters give rise to effectively the same values of local SAR and wbSAR, allowing, therefore, the same B₁ strength.
Next, the averaged E_tan to, and P_dep around, a 40-cm long generic left-side pacemaker, for a set of ten birdcages, eight anatomical models, all imaging positions from head to foot, at circular and near-circular polarizations are shown in Figure 4, as well as the results from study 2.
The results of study 3 are summarized in Figure 4. Figure 4(a) shows the B₁ field distribution over the FOV. Figure 4(b) shows the induced in vivo electrical field distribution of Duke, where birdcages of different rung numbers show very similar in vivo induced field distributions. These birdcages also have the same coefficient of variation (CoV) (23%) of the induced electrical field over the FOV. Similar results are observed for the simulated in vivo E_tan along a typical spinal cord routing in the Duke model, as shown in Figure 4(c). The observable variations in the E_tan are due to the non-ideality of the linear field polarization obtained with birdcages of varying rung numbers.

Conclusions

The evaluations performed show that birdcage diameter has only a very limited effect on the in vivo RF-induced SAR distribution, whereas birdcage length plays a critical role in determining in vivo RF-induced SAR. The sensitivity analysis of other parameters, such as high-pass vs band-pass birdcage design, birdcage current source, dorsal or lateral patient shifts, birdcage rung count, led to negligible variation compared to the dominant sources of patient anatomy, scan position, and birdcage polarization and size.

Acknowledgements

No acknowledgement found.

References

1. Nyenhuis JA. Park SM, Kamondetdacha R, et al. MRI and implanted medical devices: Basic interactions with an emphasis on heating. IEEE Trans Device Mater Reliab 2005;5(3):467–479.

2. Muranaka H, Horiguchi T, Usui S, et al. Dependence of rf heating on SAR and implant position in a 1.5 T MR system. Magn Reson Med Sci 2007;6(4):199-209.

3. Kangarlu A, Schellock F, and Chakeres D. 8.0-tesla MR system: temperature changes associated with radiofrequency-induced heating of a head phantom.

4. Murbach M, Neufeld E, Capstick M, et al. Virtual population-based assessment of the impact of 3 Tesla radiofrequency shimming and thermoregulation on safety and B1+ shimming. Magn Reson Med 2016;76:986-997.

5. Gosselin M-C, Neufeld E, Moser H, et al. Development of a new generation of high-resolution anatomical models for medical device evaluation: the Virtual Population 3.0. Phys Med Biol 2014;59:5287–5303.

6. IT'IS. https://itis.swiss/virtual-population/explib/overview/

7. IEC. IEC standard 60601-2-33. Medial electrical equipment - particular requirements for the basic safety and essential performance of magnetic resonance equipment for medical diagnosis. 2010.

Figures

(a): Generic MRI birdcage resonators and anatomical models from Virtual Population, used for study 1. (b) and (c) are additional birdcage designs generated for parameter sensitivity analysis of study 2, and 3, respectively.

Values of hSAR (red), wbSAR (blue), and pbSAR (grey) at normal operating mode as a function of imaging positions, with birdcages of bore diameter B = 70 cm and variable length L, for (a) Duke and (c) Fats; the distribution under all field polarizations are shown as shaded regions, with the circular polarization (CP) results shown as solid lines and standard limits⁷ represented by dashed lines. Allowed B₁ field strength for (b) Duke and (d) Fats are shown at normal operating mode as a function of imaging position; the distributions are shown as shaded regions, and the CP is shown as dashed lines.

Values of hSAR (red), wbSAR (blue), and pbSAR (grey) at normal operating mode as a function of imaging positions with birdcages of variable bore diameter B and length L = 70 cm, for (a) Duke and (c) Fats; the distributions under all field polarizations are shown as shaded regions, with the CP results shown as solid lines and standard limits7 represented by dashed lines. Allowed B₁field strength for (b) Duke and (d) Fats at normal operating mode are shown as a function of imaging position; the distributions are shown as shaded regions, and the CP is shown as dashed lines.

Average tangential E-field (a) and power deposition for a generic implant (b) along a generic pacemaker routing at 64 MHz under normal operating mode. Black bars represent the distribution over all birdcages at the worst-case imaging positions and its two adjacent (±10 cm) landmarks at circular polarization (CP); grey bars represent the deviation from CP of ±10°, ±20° and ±30°. Lowest and highest values for each plot are labeled. Fats Position^*: offsetting Fats by 2.5 cm laterally or 10 cm dorsally. Fats Coil/Feed^*: high-pass vs. band-pass birdcage and of leg-fed vs. IQ-fed currents.

(a) The B₁ field distribution at the thorax imaging position FOV for birdcages with rung numbers (N) of 8, 16, and 32. (b) Simulated in vivo electric field induced over the FOV for birdcages with N = 8, 16, and 32, showing negligible difference in distribution or coefficient of variation (CoV). (c) Amplitude and phase of the induced Etan along a typical spinal cord stimulator routing for N = 8, 16, and 32. Variation of the length-averaged E_tan amplitude are less than 10%.

Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)

4187