Assessing safety and transceive performance of volumetric wireless coil for wrist MRI at 3 T

doi:10.21203/rs.3.rs-1738352/v1

Download PDF

Research Article

Assessing safety and transceive performance of volumetric wireless coil for wrist MRI at 3 T

https://doi.org/10.21203/rs.3.rs-1738352/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The concept of a volumetric wireless coil electromagnetically coupled to a body coil (BC) to improve the quality and radiofrequency (RF) safety of wrist MRI at 1.5 T was recently proposed and confirmed via in vivo studies. Here, we demonstrate, for the first time, that the same concept can be applied at 3 T, where the RF safety issue is more critical. Our work includes a detailed analysis of the wireless coil robustness to the possible breakages of coil elements and its imperfect positioning inside the BC.

Methods

The wireless coil design is based on a periodic array of coupled split-loop resonators. The cases of a wireless coil breakage and inaccurate positioning were considered in terms of influence on RF safety via electromagnetic simulations with a human voxel model. In vivo experiments were also performed to evaluate the transmit efficiency and uniformity of the B₁⁺-field.

Results

Numerical simulations demonstrated that the wireless coil provides a 4.1-fold increase in the local RF safety and almost a 7-fold increase in the global RF safety in comparison with the case of the same BC used alone. Minor breakages and positioning errors only slightly affect the transmit efficiency and the RF safety, confirming the wireless coil robustness.

Conclusion

The wireless coil was shown to be feasible for wrist imaging at 3 T. A significant improvement of the RF safety may be useful in studies with implants and in the case when high-performance pulse sequences are required for improving the image quality.

MRI

Radiofrequency coils

Wireless coil

SAR

Transmit efficiency

RF safety

wrist MRI

Magnetic resonance imaging (MRI) is one of the most informative methods for visualizing manifestations of degenerative and inflammatory diseases in joints [1, 2]. Complex and thin structures in small joints, e.g., cartilage and synovial membrane, should be scanned with a very high resolution and thus require the utilization of local dedicated radiofrequency (RF) coils [3, 4]. Recently, a new concept of a dedicated volumetric wireless coil was proposed for a 1.5 T wrist MRI [5, 6]. The wireless coil was based on an array of coupled split-loop resonators [7] that focuses the transmit and receive capabilities of a whole-body “birdcage coil” (BC) within the region of interest (ROI). This concept more than doubled the signal-to-noise ratio (SNR) in comparison to a 4-channel flexible receive array coil. At the same time, it provided a dramatic decrease in the maximum local specific absorption rate (SAR) of a BC, allowing the safe utilization of efficient SAR-demanding pulse sequences [8]. Meanwhile, 3 T MRI has demonstrated advantages in joint diseases studies [9, 10]. However, SAR becomes a more critical issue at 3 T than 1.5 T due to the raise of operation frequency [11]. Thus, the aim of this work is to numerically investigate the RF safety aspects of a volumetric wireless coil for wrist MRI at 3 T. As the instabilities of these characteristics may cause severe heating of the patient’s body, we thoroughly study the coil robustness to the possible breakdowns and imperfections of the coil positioning inside the BC.

2.1 Electromagnetic simulations

All the electromagnetic calculations were performed using the CST Studio Suite 2020 software package for numerical simulations. A realistic whole-body high-pass BC model tuned and matched to 123.2 MHz was used to excite the wireless coil. The BC consisted of 16 copper legs with a length of 650 mm and had a diameter of 700 mm. An RF shield with a diameter of 826 mm and a length of 1250 mm was placed around the BC. For a quadrature excitation, two ports were used.

A previously proposed numerical model of the wireless coil for 1.5 T [6] was adapted for 3 T by tuning its lower-frequency volumetric mode to 123.2 MHz. To do this, the distributed capacitance was adjusted by decreasing the dielectric permittivity of the substrate (Arlon 25N laminate with ɛ=3.38 was substituted by Arlon 255C with ɛ=2.55) and increasing the thickness up to 2 mm and changing the copper strips size. Thus, the wireless coil consists of two dielectric substrates (dielectric loss tan δ = 0.0018) with dimensions of 244×124×2 mm³, and with copper strips metallization on the outside with dimensions of 118×3 mm² and on the inside of 42.5×10 mm², connected by 280 mm long copper tubes.

The voxel model “Gustav” (38 y male, 69 kg, 176 cm height) from the CST voxel family library was used to evaluate the wireless coil performance in the presence of a human body. The body model had a prone position with an arm slightly bent at the elbow and a wrist placed in the middle of the coil, i.e., the so-called “Superman” position. We used the Finite Integral Technique time domain solver for the simulations with a voxel model. A general view of the numerical model of the wireless coil placed inside the BC is shown in Fig. 1.

We considered two setups during the electromagnetic studies: (1) the reference case, when the BC was used alone, and 2) the BC combined with the wireless coil placed around the human model wrist. Using the MRI post-processing tool in CST, the RF magnetic field (B₁⁺) and SAR were calculated for 1 W of total accepted power. RF safety (or SAR efficiency) was calculated as the ratio of the root mean squared value (RMS) of the B₁⁺-field (averaged in a cubic area with a 7 cm edge in the wrist) divided by the square root of the maximum local SAR (local RF safety) or whole-body SAR (global RF safety).

During the study of RF safety robustness, the following cases were considered: breakage of one of the brass tubes (a tube disconnected on one side, at the soldering point to the substrate) and inaccurate coil positioning (the wireless coil rotated around the y-axis by 5°-60°). For the latest study, we used a homogeneous human body phantom with a permittivity of 70 and conductivity of 0.4 S/m to perform simulations in the Frequency Domain Solver.

2.2 Prototyping and in vivo studies

A prototype of the wireless coil was assembled using 40 telescopic brass tubes and 2 printed circuit boards (PCBs) with parallel-plate capacitors, based on the results of numerical optimization. For precise tuning to the Larmor frequency, each of the telescopic tubes had a variable length (L = 26.7–32 cm), made from two tubes of slightly different diameters of 2 mm and 3 mm, sliding against each other to create an electrical contact. A plastic case protecting the hand from contact with metallic parts of the wireless coil during scanning was designed and printed on a 3D printer.

MR studies of a healthy volunteer (22y male, 60 kg, 176 cm height) were approved by the local ethics committee and conducted in Almazov National Medical Research Center (St. Petersburg, Russian Federation). We used a 3 T Siemens MAGNETOM Trio clinical scanner. We considered two setups: (1) BC in transmit mode, with the 4-channel flex coil (from Siemens) for receiving MR signal; (2) BC in transceive mode in combination with the wireless coil.

To obtain experimental B₁⁺ maps (i.e., flip-angle maps), we applied a double-flip-angle method based on the acquisition of two single-slice gradient-echo images (TR/TE = 5000/3.7 ms, field of view = 162×399 mm², image matrix = 62×192, slice thickness = 5 mm) with flip angles of 30° and 60° for both RF transmit setups [12]. B₁⁺ maps were calculated in MATLAB 2020 (Mathworks). The B₁⁺-field inhomogeneity was estimated as the coefficient of variation (Cv), which is the ratio of the standard deviation to the mean value taken from the region of interest (ROI) – a 7×7 cm² rectangular area in the central coronal cross-section of the wrist - multiplied by 100%. A reference voltage used by the scanner to create a 180° flip angle in the central axial slice was recorded for both setups to estimate the increase of the transmit efficiency provided by the wireless coil.

To estimate the receiving performance of the wireless coil, we acquired MR images of the volunteer’s wrist using 3D gradient echo pulse sequence with fat suppression: flip angle = 10°, TR/TE = 10/3.9 ms, field of view = 100×100 mm², and image matrix = 256×256, and compared them with those obtained with a dedicated transceive extremity coil from Siemens (CP Extremity). Signal-to-noise (SNR) values were calculated on a medial slice of each MR image as a ratio of the mean signal from three ROIs in muscle tissue to the standard deviation of noise in a signal-free area.

3.1 Electromagnetic simulations

The resonant wireless coil was placed into the BC and tuned to the same frequency (123.2 MHz) that resulted in their strong electromagnetic coupling. The original resonance in the spectrum of the BC was split into two minima. Thus, in this RF setup, we observed that the peak of the RF magnetic field at 119.72 MHz provided the maximum efficiency of the wireless coil. With a real model, the frequency of the wireless coil inside the body coil can be fine-tuned by changing the brass tubes’ length.

Figure 2 shows the |B₁⁺|_RMS magnetic field and SAR maps for 1 W of total accepted power for both considered RF setups. The numerical values of the B₁⁺, and local and whole-body SAR in the ROI are presented in Table 1 together with the calculated RF safety. The transmit efficiency gain when using the wireless coil is 8.9. The wireless coil led to a 4.1-fold increase in the local RF safety, and the whole-body RF safety increased 6.7-fold.

Table 1

Numerically calculated results of transmit and SAR characteristics normalized to 1 W of total accepted power.
Coil type	\|B₁+\|_RMS, uT	psSAR_10g, W/kg	wbSAR, W/kg	\|B₁⁺\|_RMS/√psSAR, uT/√W/kg	\|B₁⁺\|_RMS/√wbSAR, uT/√W/kg
Body coil	0.36	0.23	0.007	0.75	4.27
Body coil & wireless coil	3.17	1.05	0.012	3.09	28.65

Dependence of the local and global RF safety values on the numerical position of a disconnected tube is presented in Fig. 3. We considered the cases when the 1st, 4th, 8th, 12th, 16th and 20th tubes in one of the two layers had no electrical contact with the parallel-plate capacitors. In the reference case with the normally operating wireless coil, one may see a straight line, since nothing has changed in the system. The breakage of the central tubes (especially tubes #8 and #12) had a more pronounced effect. The global RF safety was reduced 1.12-fold on average in comparison with the case of normal coil operation, while the local RF safety was improved (1.03-fold on average), even higher than in the situation of a disconnected side tube. Overall, the changes of the RF safety range ranged from 2–15%.

Figure 4 shows the B₁⁺ field distributions for the case when the wireless coil is rotated around the y-axis by α = 0°– 60° inside the body coil. The rotation of the wireless coil also resulted in the decrease of its focusing ability. Thus, the B₁⁺-field amplitude decreased, as well as the local and global RF safety characteristics (Table 2). A slight error in coil positioning (5°) resulted in a 7% decrease in whole-body RF safety and a 3% decrease in the local RF safety. The largest rotation angle (60°) more than halved both the local and whole-body RF safety.

Table 2

Numerically calculated results of the transmit and SAR characteristics normalized to 1 W of total accepted power for different positions (rotation). α corresponds to the angle between the wireless coil and the z-axis when the coil rotates around the y-axis
Coil type		\|B₁⁺\|_RMS, uT	psSAR_10g, W/kg	wbSAR, W/kg	\|B₁⁺\|_RMS/√psSAR, uT/√W/kg	\|B₁⁺\|_RMS/√wbSAR, uT/√W/kg
Body coil		0.26	0.25	0.0082	0.53	2.91
Body coil & wireless coil	α = 0°	2.46	1.05	0.0095	2.40	25.24
	α = 5°	2.43	1.08	0.0094	2.34	25.06
	α = 10°	2.40	1.12	0.0092	2.26	25.02
	α = 15°	2.35	1.15	0.0091	2.19	24.64
	α = 30°	2.12	1.24	0.0093	1.90	21.98
	α = 60°	1.11	0.96	0.0112	1.13	10.49

3.2 Experimental results

Figure 5 shows the experimentally measured flip angle maps of the healthy volunteers’ wrist obtained without and with the prototype of the wireless coil (Fig. 5A). One can observe a highly homogeneous distribution of the B₁⁺-field in the ROI (wrist joint) for both setups. Inhomogeneity of the B₁⁺-field (coefficient of variation) was calculated in the ROI indicated with solid black lines in Fig. 5B. For the BC working alone, it approached 23%, which was very similar to the results for the case when the wireless coil was combined with the BC (Cv = 22%). According to the RF power calibration procedure, the recorded reference voltage was 431.6 V when the BC was used alone and 54.7 V when the wireless coil was used together with the BC.

Figure 6 presents MR images and SNR values for the dedicated extremity coil and wireless coil. The wireless coil significantly improved the receiving capabilities of the BC, making the image quality 78% higher in terms of SNR in comparison to the state-of-the-art local coil.

This work demonstrates for the first time that a volumetric wireless coil inductively coupled to a BC can be effectively applied for the wrist imaging at 3 T with expected RF safety improvement and image quality comparable with that of a standard dedicated volumetric coil. Numerical simulation results and in vivo studies demonstrated that the wireless coil focuses the magnetic flux of the large BC within the relatively small ROI, thus, increasing the transmit efficiency more than 8-fold. A little difference between the numerically predicted 8.9-fold and the experimentally obtained 7.9-fold increase in the B₁⁺ efficiency is due to the different coil load (different hand sizes) and imperfect tuning of the wireless coil to the Larmor frequency. In the meanwhile, the B₁⁺-field distribution homogeneity in the ROI slightly improved in the experiment in the presence of the wireless coil. The obtained experimental B₁⁺ maps (flip angle maps) were in good agreement with the numerical ones, and confirmed the correctness of the RF power calibration procedure.

An introduced disconnection of one of the tubes led to the detuning of the wireless coil from the Larmor frequency. In other words, the coupling between the wireless coil and the BC became weaker, and the effect of magnetic field focusing was reduced. At the same time, the amplitude of the electric field that is mostly localized within the capacitive load also became smaller. In addition, as the wireless coil structure consists of a set of strongly coupled elements, the coupling between the middle elements is higher than that between the side ones. Thus, a broken side tube only slightly affected the resonant frequency of the wireless coil, and in these cases, the B₁⁺-field was higher than for a broken middle wire. In the meantime, in the case of a broken side tubes, the maximum local SAR was reduced faster than the whole-body SAR value. Thus, it turns out that the broken side tubes almost did not affect the global RF safety, while the local RF safety was increased in comparison with the normal operation of the wireless coil.

The wireless coil may be rotated when it is required for comfortable patient positioning. The numerical simulation results show when the rotation angle increases, the RF safety characteristics decrease. However, even when this angle is very large (60°), the RF safety characteristics remain better than for the BC working without the wireless coil, with a 2.1-fold and 3.6-fold raise in the local and in the whole-body SAR efficiency, correspondingly. Considering that in a real experiment such a big rotation does not seem likely, we may conclude that a possible minor rotation cannot cause unsafe situations. Thus, the coil RF safety characteristics are robust to minor imperfections of the experimental setup that may occur during actual scanning.

Taking into account a significant enhancement of the image quality in comparison to the dedicated commercial extremity coil, the proposed setup can be considered as safe and effective alternative. The other benefits of the wireless coil are the absence of an RF cable, i.e., less time required for coil positioning and dismounting, its light weight, robustness to possible damages, and compatibility with clinical scanners of various manufacturers. Due to the increased RF safety, the wireless coil can be potentially helpful for increasing the safety of MR scanning of patients with implants and in the situations when high-performance pulse sequences are required for improving the MR image quality [13, 14].

In the present work, we have demonstrated that a concept of a volumetric wireless coil previously proposed for 1.5 T wrist imaging can be successfully transferred to a 3 T MRI. Transmit efficiency and RF safety of a whole-body BC can be drastically improved using this wireless coil as a secondary inductively coupled element. In terms of the receive sensitivity, this setup outperforms the commercial dedicated extremity coil.

Acknowledgments

The part of the research devoted to the wireless coil design and experimental studies was supported by the Ministry of Education and Science of the Russian Federation (075-15-2021-592). The part of the work related to the RF safety assessment of the wireless coil was carried out with the support of a grant from the President of the Russian Federation for scientific school HШ-2359.2022.4.

Author contribution statement

A.N. performed the simulations; P.T. built a prototype; E.B. and A.N. performed the experiments; A.N. drafted the manuscript; P.T., A.V. and E.B. reviewed the manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

Ministry of Education and Science of the Russian Federation (075-15-2021-592); grant from the President of the Russian Federation for scientific school HШ-2359.2022.4.

A. Barile et al., Radiol. Clin. North Am. (2017) https://doi.org/10.1016/j.rcl.2017.04.006
M. Østergaard, M. Boesen, Radiol Med. (2019) https://doi.org/10.1007/s11547-019-01014-y
D. Glinatsi et al., J. Rheumatol. (2015) https://doi.org/10.3899/jrheum.141009.
C.G. Peterfy et al., Arthritis Res. Ther. (2013) https://doi.org/10.1186/ar4202
A.V. Shchelokova et al., Magn. Reson. Med. (2018) https://doi.org/10.1002/mrm.27140
E.A. Brui et al., J. Magn. Reson. (2022) https://doi.org/10.1016/j.jmr.2022.107209
S.I. Maslovski et al., Prog. Electromagn. Res. (2005) http://doi.org/10.2528/PIER04101101
E.A. Brui et al., MAGMA. (2021) https://doi.org/10.1007/s10334-021-00937-w
G. Wieners et al., Eur. Radiol. (2007) https://doi.org/10.1007/s00330-006-0539-0
M.L. Anderson et al., J. Hand Surg. (2008) https://doi.org/10.1016/j.jhsa.2008.02.028
C.K. Chou et al., Bioelectromagnetics (1996) https://doi.org/10.1002/(sici)1521-186x(1996)17:3%3C195::aid-bem5%3E3.0.co;2-z
C.H. Cunningham et al., Magn Reson Med. (2006) https://doi.org/10.1002/mrm.20896
D. Kim et al., Magn. Reson. Med. (2009) https://doi.org/10.1002%2Fmrm.22140
F. Del Grande et al., JA Radiographics. (2014) https://doi.org/10.1148/rg.341135130

No competing interests reported.

Download PDF

Editorial decision: Major revision
31 Aug, 2022
Reviews received at journal
25 Jul, 2022
Reviewers agreed at journal
09 Jul, 2022
Reviewers invited by journal
09 Jul, 2022
Editor assigned by journal
09 Jul, 2022
Submission checks completed at journal
08 Jul, 2022
First submitted to journal
08 Jun, 2022

You are reading this latest preprint version

Assessing safety and transceive performance of volumetric wireless coil for wrist MRI at 3 T

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1 Background

2 Methods

2.1 Electromagnetic simulations

2.2 Prototyping and in vivo studies

3 Results

3.1 Electromagnetic simulations

3.2 Experimental results

4 Discussion

5 Conclusion

Declarations

References

Additional Declarations

Status:

Version 1