1135

Measurement-based safety assessment, prediction and mitigation of RF induced implant heating with parallel transmission: temperature matrix

Berk Silemek¹, Lukas Winter¹, Frank Seifert¹, Harald Pfeiffer¹, and Bernd Ittermann¹
¹Physikalisch-Technische Bundesanstalt, Berlin, Germany

Synopsis

A Measurement-based Temperature-Matrix approach is presented that enables a fast, patient and exam specific estimation and mitigation of RF hazards of implants. Various locations in phantom are tested using an 8-channel (300MHz) implant safety testbed. Heating reduction Based on T-Matrix Measurements resulted >3 times heating reduction vs. circularly-polarized mode and >19 times vs. worst-case mode. 2-channel MRI (3T) feasibility experiments using high temperature resolution showed good correlation with transmitted power. In addition, T-matrix-based temperature increase predictions successfully demonstrated. As summary, an easy to implement, cheap, sensor-based method, the T-matrix, to investigate, characterize and mitigate RF heating of implants is introduced.

Introduction

Parallel transmission (pTx) is an effective method to substantially reduce RF induced heating in medical implants^1–3. Simulation and in-vitro-based mitigation approaches cannot provide a patient and exam specific assessment. Approaches that rely on integrated sensors in and around the implant allow for a patient-specific, real-time mitigation, while simultaneously the sensors provide an online safety monitoring^4–6. In this work, we introduce the temperature matrix (T‑matrix), a pTx monitoring and mitigation approach based on a single thermistor (0.1$) that can easily be integrated within the implant. This approach enables a fast, patient and exam specific estimation and mitigation of RF hazards, in principle without the need for prior simulations or in vitro testing. Feasibility is demonstrated by applying the T-matrix for heating prediction and mitigation using an 8-channel pTx coil at 300MHz in an implant-safety testbed and a 2-channel body coil inside a 3T MRI scanner.

Methods

Instrumentation: An implant safety testbed with an 8-channel 7T pTx RF coil⁶ and a cylindrical PVP based phantom is used for benchtop experiments (Fig.1a-b). A semi-rigid coaxial cable with an uninsulated tip is used as a guidewire substitute (Fig. 1b). An NTC thermistor (NCP18XH103F03RB, Murata) is connected between inner and outer conductor at the tip of the cable (Fig. 1c).
T-matrix: Tip heating impulse responses based on a complex voltage excitation vector $$$\vec{u}$$$(n) of an n-channel array are measured and used to calculate the Hermitian T-matrix $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$: $$\Delta T = \vec{u}^H\overset{\text{$\leftrightarrow$}}{T_M}\vec{u} \label{eq:Tmat} $$
To determine $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$, RF pulses with phase differences 0° (i<j) or 90° (i>j) are applied to channels i, j and the corresponding temperature rise, ΔT_ij, is recorded. In order to demonstrate a proof of concept mitigation of RF induced heating, an orthogonal projection (OP) mode was applied, which is the projection of the circular polarized (CP) vector onto the subspace orthogonal to the $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ eigenvector with the largest eigenvalue. OP allows to mitigate RF induced heating, while maintaining sufficient B₁⁺ homogeneity for imaging^6,7. The worst case (WC) is defined using the phases of the $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ eigenvector with the largest eigenvalue and equal amplitudes, which corresponds to the worst possible RF induced heating using a pTx transmission system with limited forward power per channel.
Benchtop Experiments (300MHz): Repeatability was tested with 10 $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ measurements at the same location (64 cycles: 2s RF heating/2s cooling). The T-matrix (64 cycles: 1s RF-heating/1s cooling) was evaluated on three locations within the phantom. The transmitted average power over the acquisition window was 3.14W. RF heating experiments in CP, OP and WC mode were performed with the same total forward power of 2.06W at Pos-1, 2.47W at Pos-2 and 2.41W at Pos-3 (Fig.3).
MRI Experiments (3T): To investigate the feasibility of the $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ concept under pulsed MRI conditions, a 3T scanner (Siemens, Verio) with a 2-channel body coil (quadrature splitter of the system was removed, and two ports were connected to a pTx system) was utilized Fig.4a. A conventional GRE sequence was used (TR:15ms, pulse length=2ms) to determine the relation between pulsed power vs. temperature increase at the tip.

Results

Benchtop Experiments (300 MHz): The temporal resolution of the thermistor reading was ~70ms with a temperature resolution of ~100μK. Repeatability experiments showed <3% standard deviation in slopes and eigenvalues for 10 $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ measurements (Fig.2). was acquired in 128s with a total temperature rise of <0.1°C at the guidewire tip. Based on the $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$, the calculated OP mode was able to substantially reduce RF induced heating at the guidewire tip at all three locations (>3 times vs. CP mode and >19 times vs. WC mode, Fig. 3).
MRI Experiments (3T): MR B₁⁺ image artifacts due to the induced current on the guidewire are displayed in Fig. 4b-g. The temperature increase during a 1s GRE sequence correlates well with transmitted power (Fig. 5a-b), demonstrating the feasibility of using the thermistor in MR experiments. $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ was acquired at two positions (Figs. 5c,e) in a total of 32s and with an overall temperature increase at the tip of <0.1°C. $$$\overset{\text{$\leftrightarrow$}}{T_M}$$$ contains the full system information and can also be used to predict heating. These predictions correlate well with actual measurements for different pTx driving conditions changing both amplitude (Fig. 5d) and phase (Fig. 5f).

Discussion

The T-matrix can be acquired with good repeatability and speed, requiring only low flip angles. The pTx-based method is able to detect, predict and mitigate guidewire tip temperatures as demonstrated for a 2-channel body-coil at 3T and an 8-channel pTx RF coil for 7T. Based on temperature measurements (not fields or currents), it allows to directly monitor and manage the applied thermal dose⁹. In future experiments more implant locations, implant orientations and implant geometries will be tested with the introduced method. Furthermore, B₁⁺ homogeneity will be included in the cost function to reduce tip heating.

Conclusions

An easy to implement, sensor-based method, the T-matrix, to investigate, characterize and mitigate RF heating of implants is introduced. The technique is fast and does not require additional simulations or prior in vitro testing. It is thus attractive for pTx based temperature control and monitoring of implant temperatures for patient-specific and exam-specific safety assessments.

Acknowledgements

This work was funded by the EMPIR grant 17IND01 MIMAS. The EMPIR initiative is co- funded by the European Union’s Horizon 2020 research and innovation program and the EMPIR participating states.

References

1. Eryaman Y, Guerin B, Akgun C, et al. Parallel transmit pulse design for patients with deep brain stimulation implants. Magn Reson Med. 2015;73(5):1896-1903.

2. McElcheran CE, Golestanirad L, Iacono MI, et al. Numerical Simulations of Realistic Lead Trajectories and an Experimental Verification Support the Efficacy of Parallel Radiofrequency Transmission to Reduce Heating of Deep Brain Stimulation Implants during MRI. Sci Rep. 2019;9(1).

3. Guerin B, Angelone LM, Dougherty D, Wald LL. Parallel transmission to reduce absorbed power around deep brain stimulation devices in MRI: Impact of number and arrangement of transmit channels. Magn Reson Med. 2020;83(1):299-311.

4. Etezadi-Amoli M, Stang P, Kerr A, Pauly J, Scott G. Controlling radiofrequency-induced currents in guidewires using parallel transmit. Magn Reson Med. 2015;74(6):1790-1802.

5. Silemek B, Acikel V, Oto C, et al. A temperature sensor implant for active implantable medical devices for in vivo subacute heating tests under MRI. Magn Reson Med. 2018;79(5).

6. Winter L, Seifert F, Hoffmann W, et al. Parallel transmission medical implant safety testbed: First application using time-domain E-field probes to measure and mitigate RF induced currents. Proc . ISMRM 2019:0728

7. Seifert F, Weidemann G, Ittermann B. Q matrix approach to control implant heating by transmit array coils. Proc. ISMRM 2015:3212.

8. Han H, Moritz R, Oberacker E, Waiczies H, Niendorf T, Winter L. Open Source 3D Multipurpose Measurement System with Submillimetre Fidelity and First Application in Magnetic Resonance. Sci Rep. 2017;7(1):13452.

9. Van Rhoon GC, Samaras T, Yarmolenko PS, Dewhirst MW, Neufeld E, Kuster N. CEM43°C thermal dose thresholds: A potential guide for magnetic resonance radiofrequency exposure levels? Eur Radiol. 2013;23(8):2215-2227.

Figures

Figure 1: a) Implant safety testbed consisting of an 8-ch pTx RF chain and a submillimeter positioning system^6,8b) 8-ch pTx head coil (shield removed) operating at f = 297 MHz loaded with a cylindrical liquid phantom (ε= 43.80, σ = 0.35 S/m). The testbed represents a 7T head coil and a (scaled) 3T body coil, alike. A semi-rigid coaxial cable (length 17.7cm, 7 cm immersed in liquid) with an uninsulated tip (1.5 cm of outer conductor insulation is removed). At the tip thermistor is located. c) Close-up picture of thermistor on the guidewire substitute (coaxial cable)

Figure 2: a) 10 consecutive T-matrix measurements and temperature curves of 64-cycles RF heating cycles (8x8-channel) b) derivative of the temperature from one single T-matrix acquisition from (a) to calculate the temperature slopes. c) Zoomed version of a single (2s) heating cycle illustrating the high temperature measurement precision and temporal resolution. d) Average and standard deviation of the T-matrix eigenvalues. Calculated T-matrix e) amplitudes and f) phases from a single measurement.

Figure 3: T-matrix acquisition, obtained T-Matrix magnitudes and pTx mitigation at three locations Pos-1, Pos-2 and Pos-3 within the phantom. RF heating experiments are demonstrated for CP, OP and WC mode transmission. A maximum temperature increases of 63mK(WC), 54mK(CP) and 3mK(OP) at Pos-1, 138mK(WC), 59mK(CP) and 7mK(OP) at Pos-2 and 121mK(WC), 18mK(CP) and 5mK(OP) at Pos-3 were recorded.

Figure 4: a) Experimental setup in a 3T MRI. The guidewire substitute with the thermistor at the tip was connected over a low pass filter to a coaxial cable. Outside the RF shielded room, this coaxial cable was connected to a digital multimeter in order to detect temperature dependent changes in resistance. (b-g) MR Images at the proximity of the tip for different driving conditions display the typical image artifacts due to the induced currents in the wire. The gradual change in channel weights alters the image artifact which also corresponds to different temperature increases.

Figure 5: Results from the experimental setup inside and MRI. a) Temperature rises vs. nominal flip angle from the scanner. b) Slopes of the temperature increase vs measured transmit power. c) T-matrix raw data (2_channel = 4 measurements). d) Temperature increase measured during MRI for different excitation voltages vs. T-matrix-based predictions. e) T-matrix raw data from a second location and the corresponding f) temperature increase for different excitation voltages and phases compared to T-matrix-based predictions.

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

1135