Abstract— Magnetic Resonance Imaging (MRI) access remains conditional to patients with conductive medical implants, as RF heating generated around the implant during scanning may cause tissue burns. Experiments have been traditionally used to assess this heating, but they are time- consuming and expensive, and in many cases cannot faithfully replicate the in-vivo scenario. Alternatively, ISO TS 10974 outlines a four-tier RF heating assessment approach based on a combination of experiments and full-wave electromagnetic (EM) simulations with varying degrees of complexity. From these, Tier 4 approach relies entirely on EM simulations. There are, however, very few studies validating such numerical models against direct thermal measurements. In this work, we evaluated the agreement between simulated and measured RF heating around wire implants during RF exposure at 63.6 MHz (proton imaging at 1.5 T). Heating was assessed around wire implants with 25 unique trajectories within an ASTM phantom. The root mean square percentage error (RMSPE) of simulated vs. measured RF heating remained <1.6% despite the wide range of observed heating (0.2 ℃-53 ℃). Our results suggest that good agreement can be achieved between experiments and simulations as long as important experimental features such as characteristics of the MRI RF coil, implant’s geometry, position, and trajectory, as well as electric and thermal properties of gel are closely mimicked in simulations. Clinical Relevance— This work validates the application of full-wave EM simulations for modeling and predicting RF heating of conductive wires in an MRI environment, providing researchers with a validated tool to assess MRI safety in patients with implants. I. INTRODUCTION More than 12 million people in the USA are presently carrying a form of conductive medical implant, such as a cardiac pacemaker or a neuromodulation device. More than 75% of these patients will need magnetic resonance imaging (MRI) exams during their lifetime. Unfortunately, application of MRI is highly limited for these patients due to risk of radiofrequency (RF) heating of the tissue surrounding the implant. This phenomenon, generally known as the “antenna effect”, takes place when the electric field of MRI scanner couples with the metallic leads of the medical device and amplifies the specific absorption rate (SAR) of radiofrequency energy in the tissue, potentially causing tissue burns [1]. Substantial effort has been dedicated to assess RF heating of *Research supported by National Institute of Health grants R03EB029587 and R00EB021320. Pia Sanpitak and J. Vu are with the Department of Biomedical Engineering, Northwestern University, Evanston, IL, 60608 USA and the Department of Radiology, Northwestern University Chicago, IL 60611 USA. elongated implants using a combination of full-wave electromagnetic (EM) simulations and phantom experiments [2–10]. Specifically, ISO TS 10974 technical specification describes a four-tier approach for evaluation of MRI-induced RF heating in which the last two tiers (Tier 3 and Tier 4) are applicable to electrically long wire implants (e.g., leads with length comparable to MRI resonance wavelength in the tissue). Tier 3 evaluates the lead’s transfer function⎯the RF heating response of the lead when exposed to a uniform and controlled electric field⎯ and uses it to estimate MRI-induced RF heating when the lead is exposed to any arbitrary incident electric field encountered in vivo [11]. Tier 3 approach may yield a large level of overestimation but entails less extensive simulation efforts as the lead’s transfer function can be evaluated experimentally or through reduced-size simulations [12,13]. Tier 4 approach, which is based entirely on simulations, reduces uncertainty but requires the accurate quantification of implant’s geometry and trajectory, as well as characteristics of MRI environment. To date, the majority of publications on RF heating of wire-type implants during MRI have used a Tier 3 approach. To our knowledge, there is very little data available on the accuracy of a Tier 4 approach to estimate MRI-induced RF heating of implanted leads [14]. In this work, we performed EM simulations to estimate the local SAR at tips of wire implants with various lengths and trajectories implanted in different locations inside an ASTM-type gel phantom during RF exposure at 63.6 MHz (proton MRI at 1.5 T). We used the calculated SAR in subsequent thermal simulations to predict the temperature rise in the tissue-mimicking gel at the end of 254 seconds RF exposure. We then performed experiments that matched the simulated scenarios to the best of our abilities, mimicking phantom shape and composition; wire length, trajectory, position and material; as well as MRI RF coil and imaging landmark. A total of 25 unique wire trajectories were studied. From these, 6 trajectories were used to create a calibration curve which fitted the experimentally measured RF heating, ∆  , to the simulated temperature rise ∆  . This curve was then used to predict the ∆  from ∆  for the remaining 19 trajectories. We found a good fit during the calibration phase (R 2 = 0.96), and the calibrated model predicted the experimental RF heating with high accuracy (root mean square percent error <1.6%). B. Bhusal and B.T. Nguyen are with the Department of Radiology, Northwestern University Chicago, IL 60611 USA. K. Chow and X. Bi are with Siemens Healthineers, Malvern, PA, 19355. Corresponding Author: L. Golestanirad is with the Department of Radiology and Department of Biomedical Engineering, Northwestern University, Chicago, IL, 60611 USA. Email: laleh.rad1@northwestern.edu On the accuracy of Tier 4 simulations to predict RF heating of wire implants during magnetic resonance imaging at 1.5 T Pia Sanpitak, Bhumi Bhusal, Bach T. Nguyen, Jasmine Vu, Kelvin Chow, Xiaoming Bi, and Laleh Golestanirad, Member, IEEE 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) Oct 31 - Nov 4, 2021. Virtual Conference 978-1-7281-1178-0/21/$31.00 ©2021 IEEE 4982