Report on the development and calibration of an implant safety concet in MRI, comprised of sensor-equipped smart medical implants and pTx capable MRI scanners that enable to assess and mitigate in situ RF induced implant heating

Confidentiality Status: PU - Public, fully open (remember to deposit public deliverables in a trusted repository) Deliverable Cover Sheet Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States. 1 of 33 21NRM05 STASIS D1: Report on the development and calibration of an implant safety concept in MRI, comprised of sensor-equipped smart medical implants and pTx capable MRI scanners that enable to assess and mitigate in situ RF induced implant heating Organisation name of the lead participant for the deliverable: Physikalisch-Technische Bundesanstalt (PTB) Due date of the deliverable: 30 September 2025 Actual submission date of the deliverable: 30 September 2025 2 of 33 Glossary MRI Magnetic Resonance Imaging RF Radiofrequency SAR Specific Absorption Rate DBS Deep Brain Stimulator pTx Parallel Transmission IPG Implantable Pulse Generator AIMD Active Implantable Medical Device CP Circular Polarization OP Orthogonal Projection WC Worst Case ADC Analog to Digital Converter RMS Root Mean Square 21NRM05 STASIS 3 of 33 TABLE OF CONTENTS 1 Summary ................................................................................................................................. 4 2 Introduction .............................................................................................................................. 4 3 Virtual test environment for RF sensor-equipped smart medical implants ................................ 5 3.1 Simulated sensor signals ................................................................................................... 6 3.2 Calibration of sensor signals .............................................................................................. 8 3.3 Reference Implant(s) ........................................................................................................ 10 3.4 Reference implant (ii): simulated calibration results .......................................................... 11 4 New approach for assessing uncertainties for safety-relevant parameters in pTx-based MRI systems .................................................................................................................................. 15 4.1 Experimental background ................................................................................................. 15 4.2 Q-Matrix approach ........................................................................................................... 17 5 Calibration setup for sensor-equipped smart medical implants ............................................... 21 5.1 Final calibration setup/testbed based on 8Ch pTx body coil from DKFZ ........................... 22 5.2 Final calibration setup: influence of nonlinearities............................................................. 25 5.3 Final calibration setup: results of sensor calibration ......................................................... 26 5.4 Final Calibration setup: sensor output vs B1+rms ............................................................ 27 6 Summary and conclusions ..................................................................................................... 29 7 References ............................................................................................................................ 31 8 Related other Deliverables ..................................................................................................... 33 21NRM05 STASIS 4 of 33 1 Summary Medical implants pose a safety risk for patients in magnetic resonance imaging (MRI) due to radiofrequency induced tissue heating. Current safety practices rely on simulations covering a large variety of exposure scenarios leading to overly conservative RF power limits that reduce image quality and impose a high burden on the clinics. A smart medical implant safety concept can be used for personalized safety assessments and on-the-fly monitoring of RF-induced heating, substantially improving current clinical practice. In this report a simulation-based approach was introduced to develop and build a parallel transmission (pTx) calibration system, which will allow the determination of safety-relevant RF parameters of elongated sensor equipped implants as well as the calibration of the sensor signals including the detection of possible failure scenarios. The main feature of the reported approach is the ability of the pTx based calibration and test hardware to apply a large number (> 1000) of different exposure conditions to the implant and to determine various hazard metrics in relation to B1+rms, which complies with the normative requirements of IEC60601-2-33 and ISO/TS 10974. 2 Introduction When an MR scan is conducted on a patient with an active implantable medical device (AIMD), such as a deep brain stimulator (DBS), the interaction of the RF excitation coil of the MR system with the electrically conductive wires and electrodes of the AIMD can cause substantial increases in RFinduced local SAR and respective tissue heating around the implanted electrodes.6,7 The induced local SAR is patient-, implant-, MRand exam-specific leading to complex simulation models and high uncertainties in the RF safety assessment applied in current standards8 to prevent potential tissue damage, e.g. in the brain.9 As a consequence SAR limits are substantially reduced in the presence of AIMDs, these limitations affect the imaging, thus, diagnostic performance of the MRI system. In Deliverable D2 the smart medical implant safety concept was already introduced which is based on personalized measurements rather than generalized simulations1–4. Small and inexpensive physical sensors can be attached to the implant electrodes2, embedded in the implant casing3 or even unmodified commercial components, such as shown for DBS implants15,16, can be used to 21NRM05 STASIS 5 of 33 assess the momentary and patient-specific RF hazard. If the MR system and RF coil support more than one independent RF transmission channel, i.e. so-called parallel transmission (pTx), the sensed implant information can be intelligently utilized to substantially mitigate the RF-induced heating without the need to reduce the transmitted RF power and compromise image quality2–4,17. Suitable pTx-based testing and calibration procedures are essential for the subsequent application of the safety concept described above. Compared to established implant testing procedures for conventional MR systems with a single RF transmission channel, the large number of degrees of freedom in a pTx system leads to significantly greater complexity. On the other hand, pTx-based test systems enable a much larger number of exposure scenarios to be generated in a very short time, e.g., 1000 field configurations in 100 ms. This in turn enables stochastic test procedures and thus both the reliable identification of potentially dangerous situations and a simpler determination of uncertainties in the system. Such pTx-based test systems could therefore also be advantageous for testing implants without built-in sensors and for passive implants. A crucial point here is that the test system can generate traceable electromagnetic field configurations that reflect the typical exposure conditions in 1.5 T or 3 T whole-body MR scanners. This is achieved in the setup described here by using a body coil with 8 transmit channels, which is geometrically equivalent to the body coil of a conventional 3T MR scanner (see Deliverable D3). This allows the RF exposure to be related to the B1+rms in circularly polarized mode (CP) in accordance with the standard. The traceable determination of B1+rms at the center of the coil in pTx mode is an important result of the project. This allows safety-relevant parameters, such as the temperature rise at the implant tip, to be measured for a variety of exposure conditions in the sense of an end-to-end test. With these data, both implant and MR manufacturers can develop optimization strategies to minimize the risks of RF-induced tissue heating while improving MR imaging performance. 3 Virtual test environment for RF sensor-equipped smart medical implants 21NRM05 STASIS 6 of 33 In a first step a simulation-based approach was introduced to evaluate the characteristics of a parallel transmission (pTx) calibration and test system, which will allow the determination of safety-relevant RF parameters of elongated sensor equipped implants as well as the calibration of the sensor signals including the detection of possible failure scenarios. The virtual test environment consists of a series of model calculations that can be used to determine safety-related parameters, such as the point SAR and the local temperature rise as a function of the sensor signal from the smart implant. This allows strategies for calibrating the sensor signals to be developed and possible error scenarios due to interference and signal loss to be identified. This would enable the implant manufacturer to perform a simulation-based risk assessment. The model calculations focused on the use of RMS sensors for pTx mitigation, which proved to be the most affordable option in the previous project.1–4 3.1 Simulated sensor signals The AIMDs primarily considered are neurostimulators (DBS), in which a thin (< 2 mm) electrode measuring 30 cm to 50 cm in length is equipped with a series of electrode pads. The electrodes, which are usually shielded, have corresponding wires inside that connect to the implantable pulse generator (IPG), although the connection to the IPG is typically not completely RF shielded. The RF signals from the tip of the implant conducted via the wires may therefore be affected by RF interference in the area of the IPG and therefore only reflect the conditions at the tip of the implant to a limited extent. In addition, an RF signal is conducted to the IPG from each electrode pad in principle, and the SAR distribution at the various electrode pads may also vary. It is therefore very important to clarify in advance of further developments whether the sensor-based safety concept described in D2 still works in the case of a real implant with significantly higher complexity. To this end, simulation calculations were carried out on a simplified model of a shielded DBS electrode with 4 internal leads and 4 pads at the tip (see Fig. 1). The IPG, including the connections, was modelled on real AIMDs. The RF excitation is realised via 16 loop coils, each of which was assigned to an RF channel. The FDTD simulation was carried out with an isotropic mesh width of 0.5 mm. From the 4 different sensor signals measured inside the IPG, a linear superposition can 21NRM05 STASIS 7 of 33 then be determined in the sense of a best fit, which correlates best with the maximum point SAR, e.g. at the location of electrode pad #1 (s. Fig. 2) Figure 1 Simulated setup with 16 excitation coils and 4 electrode pads. Figure 2 Correlation of simulated hazard measure (point SAR at electrode #1) with best fit of sensor signal surrogate. As a result, there is certainly an increased uncertainty in the determination of the point SAR at the electrode pads for the selected model configuration, which is caused by EMF interference. This should be taken into account when designing sensor-equipped smart implants. In any case, the simulation calculations presented allow such effects to be analysed in advance of implant development. However, there are also other measurable variables that can be used as sensor 21NRM05 STASIS 8 of 33 signals, e.g. changes in impedance between different electrode pads15,16, which are differently sensitive to interference. In the end, the reliability of the considered sensor-equipped smart implant can only be evaluated experimentally with a pTx-based testbed. 3.2 Calibration of sensor signals For the presented safety concept to be applicable, the recorded sensor signals must be as clearly related as possible to a hazard metric. This allows this hazard metric to be recorded over time before or during an MR examination and risk-minimising measures to be taken. The most relevant hazard metric is the temperature rise at the implant tip. If the temperature rise is limited to 1K or 2K, for example, the risk of possible tissue damage can be estimated and taken into account in the implant manufacturer's risk management. It is therefore necessary to carry out appropriate simulations of the relationship between the sensor signal and the temperature increase at the implant tip for the simple reference implants considered in this project in order to identify possible systematic sources of error. Furthermore, these simulations also allow the measurement setups for the calibration measurements to be optimised, especially with regard to the positioning of fiber optic temperature probes. In Fig. 3, for example, it was shown for the reference implant ultimately selected, based on a semirigid coaxial cable, that the temperature curves in the proximity of the implant tip can be matched very well with the simulated curves if the measurement position is chosen appropriately. 21NRM05 STASIS 9 of 33 Figure 3 Simulation of temperature rise (inhouse code21). (A) Simulation setup of the coaxial cable tip. (B) Simulated normalized temperature in the plane of the coaxial cable. (C) Normalized temperature difference as a function of time for position A and B compared to a measured curve of position A. In further simulations, the effect of different tissue parameters on the temperature curves was analysed in comparison to the PVP-based phantom fluid (PVP - polyvinylpyrrolidone) used in the experimental test setup. The comparison considers the tissue types 'white matter', 'grey matter' and 'cerebrospinal fluid' relevant for DBS electrodes. Fig. 4 and Fig. 5 summarise the results for 3T and 7T respectively. These simulations allow conclusions to be drawn about the real conditions in vivo and would have to be carried out by the implant manufacturer in accordance with Tier 3 ISO/TS 10974 anyway. Figure 3: Simulation results (Sim4Life) of (A-D) RF induced E-fields of a reference implant electrode for a 128MHz RF exposure and different background tissue types and (E-H) corresponding thermal simulations at the electrode. (I) Temperature evolution over an RF heating period of 60s at the location marked with ‘x’. 21NRM05 STASIS 16 of 33 conditions in these head coils are not suitable for widely applicable test hardware, which must be based on standard MR systems (1.5T and 3T with body coil). Figure 12: (A) Correlation of the sensor signal with radial E-field for 1000 random pTx voltage vectors. pTx voltage vectors marked with a diamond were scaled by a factor of 5 and are used in (B). (B) Correlation of squared coaxial sensor signal with temperature rise after of heating for 10 random pTx vectors. Figure taken from Ref.4. Figure 13: E-field and temperature rise calibration setup used to determine correlation of the sensor signal with radial Efield and with temperature respectively. More details about the setup, field and temperature probes can be found in the corresponding references1,3,4. Figure taken from Ref.3. 21NRM05 STASIS 17 of 33 4.2 Q-Matrix approach The crucial point in the development and calibration of an implant safety concept for smart implants equipped with sensors in connection with pTx-capable MR scanners is the implementation of an affordable method for determining uncertainties that adequately takes into account the high number of degrees of freedom resulting from the use of a pTx system. Another point is that MR-compatible implants, i.e., medical devices labelled “MR conditional,” are now mostly labelled with regard to B1+rms, as specified in the latest versions of IEC 60601-2-33 and ISO/TS 10974, especially if MROC (MR Equipment Output Conditioning) is used in the future5,8. An approach is therefore being sought that will also allow the B1+rms to be determined in the test setup. For this reason alone, only a pTx body coil can be considered as the exposure system for further developments. Such a coil is described in Deliverable D3, of the STASIS project and was used in further work. We will now consider the following relevant parameters: (i) sensor signal, (ii) E-field/SAR at the tip of the implant, (iii) B1+rms at the center of the body coil, (iv) temperature increase at the tip of the implant, all of which are related to quantities that result from the superposition of the fields generated by the pTx coil. These quantities can all be represented by a corresponding quadratic form, the so-called Q-matrix QX, where X is related to quantities (i) to (iv). QX is a positive definite Hermitian matrix, meaning it has only positive eigenvalues. QX is defined locally for the measured variables under consideration, i.e., only one eigenvalue should be different from zero, and the associated eigenvector represents the maximum constructive interference of the complex-valued field variables, including the phase angle of the individual RF channels in relation to a selected channel (e.g., channel #1). From measured QX matrices, it is therefore also possible to infer the phase errors of the pTx system without using a phase-sensitive receiving system with the need to have a reference phase available. Hence, only the RMS values need to be recorded, which considerably simplifies measurement data acquisition. This means that even for phase-dependent variables, such as the induced voltage at a pick-up coil (PUC) used for B1 measurement or the voltage at the implant sensor, only RMS values need to be determined. The phase information is provided solely by the coherent pTx system. 21NRM05 STASIS 18 of 33 Therefore, the Q-Matrix QX contains all relevant information about the pTx system itself, which is why this approach enables end-to-end testing. Mathematically, this can be summarised as follows: X= u|QX|u (1) For sensor signals that are represented by a complex value Us, e.g. the RF voltages measured at the implant sensor or the B1 pick-up coil, X would then be equal to |Us|2. For SAR-related quantities or temperature-related values, X would be e.g. equal to the SAR value or ΔT, respectively. The voltage vector |u denotes the complex valued RF excitation signal setting in the pTx MRI system. The rank of QX is determined by the number of independent RF channels Nchan. Based on these properties of QX, the following procedure is used to determine quantities (i) to (iv) and the associated measurement uncertainties. 1. Measurement of QX with a simple pTx sequence: Nchan2 measurements are required to construct QX as follows: 2𝑄𝑋,𝑘𝑙={(𝑋𝑘𝑙− 𝑋𝑘− 𝑋𝑙)+𝑗(𝑋𝑘𝑙 †− 𝑋𝑘− 𝑋𝑙) for 𝑘≠𝑙 𝑎𝑛𝑑 𝑘<𝑙 (𝑋𝑘𝑙− 𝑋𝑘− 𝑋𝑙)−𝑗(𝑋𝑘𝑙 †− 𝑋𝑘− 𝑋𝑙) for 𝑘≠𝑙 𝑎𝑛𝑑 𝑘>𝑙 2𝑋𝑘 for 𝑘=𝑙 , (2) where 𝑋𝑘 is the signal X when only channel 𝑘 transmits, 𝑋𝑘𝑙 denotes the signal X when channels 𝑘 and 𝑙 transmit simultaneously in phase, and 𝑋𝑘𝑙 † the same for a 𝜋/2 phase difference between both channels, while the channels’ amplitudes are always identical. 21NRM05 STASIS 19 of 33 2. Approximate representation of QX by the eigenvector vector |evXmax at the largest eigenvalue λXmax. QX can be represented as follows: QX = λXmax |evXmaxevXmax| (3) Then the following results for any pTx voltage vector |u for the complex valued voltage Us: Us = (λXmax )0.5 evXmax|u (4) 3. Uncertainties via stochastic approach A stochastic approach is presented to determine the uncertainties in the system, both the systematic deviations, e.g. those caused by non-linearities, and the random errors (s. Fig.14,15). For example, 1000 random pTx voltage vectors are generated, whereby the maximum amplitude across all transmit channels is limited to a specified value. A sensor value can then be predicted from (4) based on the measured Q-matrix QS. This value can then be compared with the actual measured sensor values, which provides information about the uncertainties in the system. Thus, deviations from the predicted values form a genuine metric for the uncertainties in this multidimensional pTx system. 21NRM05 STASIS 20 of 33 Figure 14 Scatter plot of the predicted versus measured sensor signal for 1000 randomly selected voltage vectors (7T head coil, s. Fig. 13). When using the eigenvector with the highest eigenvalue for Qs, the correlation is better and there are fewer outliers. Figure 15 Scatter plot of the predicted versus measured E-Field at the implant tip for 1000 randomly selected voltage vectors (7T head coil, s. Fig 13). By using the eigenvector with the highest eigenvalue for Qs, some systematic errors can be avoided, especially for low sensor signals. 21NRM05 STASIS 21 of 33 Based on this methodology, the final calibration and test setup was developed. 5 Calibration setup for sensor-equipped smart medical implants The individual components of the setup are described below. The 8ch pTx body coil (Deliverable D3, Fig.16) is driven by the testbed setup from Ref.1 (see also https://www.opensourceimaging.org/project/ptx-implant-safety-testbed/). The amplifiers used allow a maximum output power per channel of 20W, whereby the linearity of the amplifiers is satisfactory for output powers of up to 10W. With this setup, up to approx. 100W total power can be applied to the pTx body coil for heating experiments over a longer period of time, which also results in easily measurable temperature increases at the implant tip. The phantom used is based on the ASTM phantom (Fig.17), but is shorter in order to reduce its mass, which greatly facilitates measurements. In principle, however, the pTx body coil is also suitable for the ASTM phantom. The reference implant is positioned in the phantom, which is filled with a PVP-water mixture3. For this purpose, a mechanical holder was 3D printed to fix the implant in position at the tip and simultaneously accommodate the sensing fiber3 of the FBG temperature instrument (Fig.18). At the end of the coaxial cable used as the reference implant, an SMA connector is attached to the cover, giving the setup the necessary mechanical stability, which is essential for reproducible measurements. The determination of B1+rms in the centre of the pTx body coil is essential for the end-to-end tests. For this purpose, a mechanical holder for the OSI2 pick-up coil was 3D printed (Fig.19), which allows the precise and reproducible fixation of the PUC in two mutually perpendicular positions. This allows the field amplitudes B1x and B1y to be determined, from which B1+ and B1are then derived. B1+ or B1in CP+ or CPmode is selected as the reference value. The specific direction of rotation of the corresponding B1 field depends on the orientation of B0. Therefore, implants should always be tested in both directions of rotation, as the geometric conditions are not necessarily symmetrical. Determining the B1 field components requires in-situ calibration in a TEM cell22, a calculable field source in which the applied voltage is derived from the measurement of the RF power used. This 21NRM05 STASIS 22 of 33 power is determined using a highly accurate calibrated power measurement head (R&S NRP18TN), which ensures the traceability of the measurements. The FBG sensors used for temperature measurements were also calibrated using a traceable PT100, which means that the determination of temperature-related hazard metrics is also traceable in principle. As it later became apparent, nonlinearities have a significant impact on the determination of the QX matrix. These nonlinearities can originate from both the RF power amplifiers and the signal acquisition of the sensors. For this reason, the primary measurements were performed with the reference implant connected directly and not with fiber optic transducers or wireless transmission of rectified sensor signals. Nevertheless, the setup can also be used for testing off-the-shelf implants if an E-field sensor or an isotropic SAR probe is placed at the tip of the implant, allowing for sufficiently fast readout. This can then also be used to calibrate a sensor signal with respect to the selected hazard metric16. 5.1 Final calibration setup/testbed based on 8Ch pTx body coil from DKFZ The following figures (Fig.16-19) visualize all hardware components used in the calibration procedure. 21NRM05 STASIS 23 of 33 Figure 16 8-channel pTx body coil used for the calibration and testing setup together with PVP phantom, reference implant inside the phantom, tip holder with temperature sensing fiber (FBG). At the end of the coaxial cable an SMA connector is attached to the cover and connected to the receive system (blue cable). The mechanical holder for the OSI2 pick-up coil is seen in the center of the coil Figure 17 Detailed view of the phantom with reference implant (red lead), tip holder with temperature sensing fiber), SMA connector and OSI2 pick-up coil. 21NRM05 STASIS 24 of 33 Figure 18 More detailed views auf the components used, upper row left: overview as seen in Fig. 17, upper row middle: detailed view of tip holder with temperature sensing fiber, upper row right: CAD model of PUC mount for TEM cell, lower row left: CAD model of tip holder, lower row middle: CAD model of holder for the OSI2 pick-up coil, lower row right: detailed view OSI2 pick-up coil with holder. Figure 19 Components of the TEM calibration setup. Left: PUC in mount for TEM cell, middle: MR-TEM cell22 with PUC inserted, right: used calibrated power measurement probe head (R&S NRP18TN). 21NRM05 STASIS 25 of 33 5.2 Final calibration setup: influence of nonlinearities Fig. 20 shows examples of results for the implant sensor at different power levels. The attenuation level att=0.3 corresponds to a maximum forward power of approximately 1W per channel, whereby it should be noted that this power is not measured directly at the coil because, in the sense of an end-to-end test, all measured quantities are related to the B1+rms at the center of the coil. CP+ and CPmodes are particularly highlighted pTx voltage vectors, whereby the implant sensor signal is very different for both modes. This emphasizes that tests must be performed in both modes. The influence of the nonlinearities of the RF power amplifiers is evident; the corresponding uncertainties (s. Fig. 21) would have to be taken into account in a real pTx system. Figure 20 Final calibration setup: results for the reference implant sensor signal for different power levels. 21NRM05 STASIS 32 of 33 10. Boutet A, Chow CT, Narang K, et al. Improving Safety of MRI in Patients with Deep Brain Stimulation Devices. Radiology. 2020;296(2):250-262. doi:10.1148/radiol.2020192291 11. Shetty AS, Ludwig DR, Andrews TJ. Invited Commentary: MRI in Patients with Active Implanted Medical Devices: Demand Will Only Grow. RadioGraphics. 2024;44(3):e230231. doi:10.1148/rg.230231 12. Pitman BM, Ariyaratnam J, Williams K, et al. The Burden of Cardiac Implantable Electronic Device Checks in the Peri-MRI Setting: The CHECK-MRI Study. Heart Lung Circ. 2023;32(2):252-260. doi:10.1016/j.hlc.2022.10.005 13. Celentano E, Caccavo V, Santamaria M, et al. Access to magnetic resonance imaging of patients with magnetic resonance-conditional pacemaker and implantable cardioverter-defibrillator systems: results from the Really ProMRI study. EP Eur. 2018;20(6):1001-1009. doi:10.1093/europace/eux118 14. Falowski S, Safriel Y, Ryan MP, Hargens L. The Rate of Magnetic Resonance Imaging in Patients with Deep Brain Stimulation. Stereotact Funct Neurosurg. 2016;94(3):147-153. doi:10.1159/000444760 15. Silemek B, Yalaz M, Seifert F, et al. Built-in RF safety for active implants: Harnessing impedance measurements from a commercial deep brain stimulator. In: Proc. Intl. Soc. Mag. Reson. Med. Vol 32. Singapore; 2024:0786. 16. Silemek B, Seifert F, Brühl R, Ittermann B, Winter L. Impedance measurements detect the temperature rise near DBS. In: Proc. Intl. Soc. Mag. Reson. Med. Vol 33. Hawai’i, United States; 2025. 17. 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):2824-2832. doi:10.1002/mrm.26914 18. Silemek B, Seifert F, Petzold J, Brühl R, Ittermann B, Winter L. Wirelessly interfacing sensorequipped implants and MR scanners for improved safety and imaging. Magn Reson Med. 2023;90(6):2608-2626. doi:10.1002/mrm.29818 19. Physikalisch-Technische Bundesanstalt (PTB). Wireless Reference Implant. Open Source Imaging. https://www.opensourceimaging.org/project/wireless-reference-implant/. Published 2023. Accessed March 27, 2025. 20. ASTM F2182 - 11. ASTM F2182 - 11 Standard Test Method for Measurement of Radio Frequency Induced Heating Near Passive Implants During Magnetic Resonance Imaging. https://www.astm.org/DATABASE.CART/HISTORICAL/F2182-11.htm. Accessed March 22, 2020. 21NRM05 STASIS 33 of 33 21. Seifert F, Weidemann G, Ittermann B. Correlation of psSAR and tissue specific temperature for 7T pTx head coils - a large scale simulation study. In: Proc. Intl. Soc. Mag. Reson. Med. Vol 23, 380. Toronto, Canada; 2015. 22. Klepsch T, Lindel TD, Hoffmann W, Botterweck H, Ittermann B. and Seifert F. Calibration of fiberoptic RF E/H-field probes using a magnetic resonance (MR) compatible TEM cell and dedicated MR measurement techniques. Biomedical Engineering/Biomedizinische Technik, vol. 57, no. SI1-Track-B, 2012, pp. 000010151520124428. https://doi.org/10.1515/bmt-2012-4428 8 Related other Deliverables D2: Report on the technical specifications and the communication workflow for sensor-equipped smart medical implants within an MRI scanner, https://doi.org/10.5281/zenodo.15502183 D3: Report on the development of open-source reference hardware (RF coil, exciter, modulators, RF power amplifier) and open-source control software, including traceable measurement procedures that allow testing of implants under pTx MR conditions