A Simulated Comparison of Local SAR Near Orthopedic Implants at 7T Relative to Lower Field Strengths
Presentations
TH-C-TRACK 3-6 (Thursday, 7/29/2021) 1:00 PM - 2:00 PM [Eastern Time (GMT-4)]
Purpose: Electromagnetic (EM) simulations were used to investigate the maximum local 1g-average SAR induced in tissue adjacent to orthopedic screws of varying lengths and field strengths.
Methods: EM simulations were performed within a finite-difference time-domain solver package (Sim4Life, ZMT, Switzerland). Orthopedic screw models, ranging in length from 18-183mm, were placed within a tissue region assigned to dielectric properties of bone. A dipole antenna, rather than a birdcage-type RF coil model, was used to create uniform background electric fields in the vicinity of the implant for each field strength ranging from 3T to 7T. Simulations were performed using a dedicated GPU (Tesla V100, NVIDIA, USA), each taking approximately 4 hours to run. Two normalization methods were implemented on simulation input power. The first accounted for reflection losses and ensured identical transmit power (1W) into the tissue volume. The second ensured a uniform B1+ magnitude (0.45μT) was achieved at the implant location. Simulations were also performed on tissue with dielectric properties of muscle and white matter brain tissue at 7T.
Results: After normalizing to a constant B1+ field, maximum 1g-average SAR as a function of screw length showed a peak near 4T; the value at 7T was 36% lower than the peak. The implant length at which the maximum SAR was measured for all field strengths occurred at approximately one-third of the wavelength at each field strength, as expected. The maximum SAR shifted to longer implant lengths in bone compared to muscle and white matter, reflecting the lower permittivity and conductivity in bone.
Conclusion: For the specific set-up investigated, a low maximum SAR value was determined at 7T compared to lower field strengths when normalized to the B1+ field. These results indicate RF-induced tissue may not necessarily be significantly higher at 7T compared to lower fields, as conventionally assumed.
Funding Support, Disclosures, and Conflict of Interest: One author (PSJ) is in receipt of support from NIH NIGMS 5 R25 GM075148
Handouts
Keywords
Taxonomy
IM- MRI : Safety, Risk evaluation & control
Contact Email