WP1: Model development
The aim of this work package is to develop anatomically correct digital models of the body of a human subject carrying medical implants, suitable for numerical EMF modelling purposes. The high (to 1 mm) resolution digital models developed will be utilised in WP2 and WP4, in order to simulate potentially hazardous tissue heating due to energy deposition into the implant and in the surrounding tissues during MRI. Active (e.g. pacemakers) and passive medical implants will be investigated, but only the “passive aspects” of the implants will be considered. In particular, models will be developed for the following three categories of medical implants:
- massive orthopaedic implants, like hip, knee and shoulder arthroplasty;
- long one-dimensional implants, like leads and electrodes of pacemakers and stimulation devices;
- small stand-alone implants with the largest extension of 10 cm, like copper Intrauterine Device (IUD),
- stents, aneurism clips, screws, plates and other fixation elements.
WP1 activities will be focused on: 1) analysis and selection of the most representative combinations of medical implants and human types (sex, age, size) with respect to the hazards generated by RF fields in the ~100 MHz regime and by switched gradient fields in the ~1 kHz regime; 2) definition of the high resolution anatomical models and tissue properties that will be used for the models; 3) collections of the properties to characterise the investigated medical implants; 4) development of the high resolution, anatomical virtual models of human subjects carrying realistic medical implants.
The WP1 target is to develop representative models for each type of medical implant (orthopaedic, long onedimensional, small implants). In total, at least twelve device models will be considered.
WP2: Numerical modelling of RF power deposition and associated temperature increase
The aim of this work package is to assess the locally deposited RF power around the identified implants of WP1, and to estimate the resulting temperature increase. In accordance with ISO/TS 10974, the incident field of the MRI RF exposure (without implant) will be assessed separately from the field induced by the implant, as the evaluation of all implants in all patients and positions is exceeding any computational possibilities.
To gain knowledge of the induced field distributions in clinical MRI scanning, up to 2560 realistic MRI exposure scenarios will be evaluated, to capture the statistical distribution and worst-case exposure for all major MRI scanner geometries and MRI patient population anatomies. Automated evaluation routines will be developed, to extract the relevant fields at the location of the implants from this field-database, and gather the statistical data.
The estimated fields will be applied to the identified implants, analysing the effect of their geometry on the additional exposure in their vicinity. Dedicated approaches will be developed and refined for both, small
implants (e.g., orthopaedic, stents) and large implants (e.g., Active Implantable Medical Device (AIMD) leadwires), as with increasing size, the phase-variation of the incident field needs to be considered.
Finally, numerical and experimental validation studies will be performed, which will help to confirm the uncertainty budget associated with each evaluation step. Selected scenarios will be simulated in full (MRI RF
exposure with the implant present in the patient), and compared to the results from the separated approach.The experimental validation will include measurements of the deposited power and temperature increase inside a bench-top RF exposure system.
The developed tools, methodologies and results will be applied in WP3 and WP5.
WP3: Active mitigation of RF induced heating of metallic implants by parallel transmission
The aim of this work package is to develop and validate a metrologically sound methodology to exploit the multiple degrees of freedom of pTx systems to actively steer radio frequency E-fields away from the metallic
implant while simultaneously maintaining image quality at an acceptable level. This includes the determination of tissue heating related quantities like local SAR (psSAR), steady state temperatures or thermal doses (ThD) in dependence of the RF steering parameters of the pTx system.
To this end model implants with either embedded or external sensors will be used to measure safety relevant parameters like temperature rise, induced RF currents or E-fields along the implant. An appropriate
experimental validation of numerical implant models within the pTx MRI system by calibrated EM field measurements will be a key result of this work package allowing a more reliable safety assessment. Further, an interface will be developed which allows a real-time feedback of such sensor measurements to actively control the heating related hazards by adjusting the pTx settings of the MRI scanner. This will result in the development of a safety concept for active E-field control in the presence of metallic implants using pTx where the mutual responsibilities of implant manufacturers and MRI manufacturers are not only clearly defined but also clearly separated from each other.
The methodology used in this work package is based on advanced simulation methods and calibrated in-situ measurements of RF EM fields near or around metallic implants within an MRI system. For the development and validation of pTx based mitigation of implant heating a bench top pTx system at PTB will be used allowing RF field probe and thermal measurements in a fully controlled environment. This measurement setup will be complementary to existing pTx capable MR scanners at PTB.
WP4: Implant related hazards due to switched gradient fields
The aim of this work package is to assess quantitatively the specific hazards, for patients carrying metallic medical implants, associated to the low-frequency magnetic fields in the kilohertz regime produced by MRI
gradient coils (GC). The adopted methodology will be based on a combination of simulations and validating experiments, taking advantage of the output from WP1 and the advanced computational tools developed in a previous EMRP project (EMRP-HLT06), available within the consortium.
The activity will be mainly focused on the heating of the implant due to the energy deposition within the device itself, and its correlated effects on the surrounding human tissues. The investigation will also explore possible implications for localised peripheral nerve stimulation (PNS) enhancement. This responds to a stakeholder request from ISO 10974 JWG, since so far no investigation of this possible risk mechanism exists. Potential safety issues related to device malfunctioning (EMC problems) for active implants are outside the scope of this project and will not be considered.
Although the adopted method of analysis could be applicable to several classes of implanted devices, in this WP the attention will be focused on orthopaedic implants & prosthetics (such as hip implants, artificial knees, etc.). This is based on estimations of the relevance of GC field effects for different types of implants. Model systems shall be looked at to unravel the underlying mechanisms and then real life implants shall be investigated to ensure practical relevance. The analysis will be carried out referring to real MR scanners (at PTB) and considering both clinical and dedicated test GC sequences. Worst case scenarios will be explored with test sequences and critical regimes with respect to gradient strengths and switching frequencies shall be identified for test and for clinical sequences. Potential mitigation solutions to limit harmful effects will be analysed and tested. The relevant parameters to quantify the hazard associated with gradient heating of implants, e.g. |dB/dt|rms, the root-means-square temporal average of the magnetic field change, will be identified and limiting values will be suggested.
The results of this WP will provide the basis for proposals of revision of laboratory testing procedures to certify MR compatibility of implants. The activities in this WP address Objective 4 of the project, that is: (a) to investigate numerically and experimentally the hazards associated with the interaction between bulk metallic implants and switched magnetic fields in the kilohertz regime; (b) to develop a reference set-up for testing metallic implant heating, using switched magnetic-field gradients of a few mT/m with a target gradient-uncertainty below 5 %.
WP5: Statistical and parametric analysis of implant heating during MRI
The aim is to allow more efficient and less time-consuming evaluations of MRI-induced heating of small medical implants by providing a parametric analysis of the implant-related factors affecting tissue heating (both RF and gradient-field heating) and a statistical analysis of the likely exposure to the regions occupied by implants for a wide range of MRI and patient scenarios. Tissue composition and thermal properties in these regions will also be assessed in order to establish correction factors for temperature rises obtained by measurements in non-perfused phantoms. This will allow accurate (and not overly conservative) thresholds to be established for different implant types within the tiered evaluation method proposed. The feasibility of providing stochastic risk assessments for implant bearing patients undergoing MRI will be assessed.
In order to evaluate efficiently the MRI-induced heating of implants, it is necessary to understand the effect of the main implant-related factors on tissue heating. This understanding helps in identifying a subset of worstcase configurations for evaluation and potentially allows the uncertainty in using results from simplified models or similar devices to be established. This work package will go beyond the state of the art by providing a parametric analysis of the implant-related parameters for the implants identified in WP1, with respect to RF and gradient induced heating in the phantom.
The work uses the software tools, models and large sets of simulations developed in WP2 and WP4 to provide the datasets for the analysis of frequency-distributions of exposures for different implant types within human surrogates, and these data sets span the range of patients, commercially available coil types and patient positions during the scan. Data will also be compiled on the distribution of whole-body SAR and total absorbed energy during the scan. Note that the prevalence of each coil type, sequence and patient position would be required to determine the frequency distribution for patient exposures during MRI.