Hanno Homann¹, Peter Börnert², Kay Nehrke², Holger Eggers², Olaf Dössel¹, and Ingmar Graesslin^2
1Institute of Biomedical Engineering, Karlsruhe Institute of Technology, Karlsruhe, Germany, ²Philips Research Europe, Hamburg, Germany

Estimation of the specific absorption rate (SAR) is typically performed by numerical simulations using generic body models. However, this represents the local SAR in the individual patient only to a limited extent. This study evaluates a recently proposed approach for generating patient-specific body models based on whole-body water-fat-separated MR data. A comparison of measured and simulated B1-fields showed qualitative and quantitative agreement. Local SAR values vary significantly between patients. However, similar locations of SAR hotspots were observed. These results increase confidence in the validity of simulated SAR values.

Andreas Klaus Bitz^1,2, Oliver Kraff^1,2, Stephan Orzada^1,2, Stefan Maderwald^1,2, Irina Brote^1,2, Sören Johst^1,2, and Mark E. Ladd^1,2
1Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Germany, ²Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany

To guarantee safe operation of RF transmit coils during MR examinations, the maximum permitted input power at which SAR limits are not exceeded must be determined. In particular, SAR estimation in high-field MR demands RF simulations with detailed numerical models including transmit coil and heterogeneous human bodies. For validation of the numerical results, an iterative procedure for RF safety testing of transmit coils is proposed which incorporates validation by field measurements in well-defined, canonical test configurations. The proposed test procedure has been applied for safety tests of local and volume coils at 7 Tesla. Results are presented for an 8-channel head and an 8-channel body coil.

Leeor Alon^1,2, Cem Murat Deniz^1,2, Daniel K Sodickson^1,2, and Yudong Zhu^1,2
1Center for Biomedical Imaging, Department of Radiology, NYU School of Medicine, New York, NY, United States, ²Sackler Institute of Graduate Biomedical Sciences, NYU School of Medicine, New York, NY, United States

Since the local specific absorption rate (SAR) is difficult to measure in vivo, one approach to ensure patient safety utilizes simulation software such as xFDTD to compare simulated and measured |B1+| fields, under the assumption that a high correlation between the two implies that the simulated local electric fields also align with the unmeasured experimental local electric fields. The numerical simulation results in this abstract indicate, however, that substantial variations in electric field and SAR are indeed possible with minimally-varying |B1+| in the setting of distinct tissue properties at 7T field strength.

Xiaoping Wu¹, Sebastian Schmitter¹, J. Tian¹, J. T. Vaughan¹, Kamil Ugurbil¹, and P-F. Van de Moortele^1
1CMRR, Radiology, University of Minnesota, Minneapolis, MN, United States

Recently, there has been an increasing interest in body imaging at 7T. One critical issue to be addressed in those applications is the severe transmit B1 (B1+) inhomogeneity present at such high magnetic field. Parallel transmission (pTX) shown to be able to reduce B1+ inhomogeneity holds great potential for body imaging. However, pTX at 7T involves complicated behavior of specific absorption rate (SAR) as was shown in the human brain. Here, we investigated SAR behavior of pTX when using 3D spoke RF pulses for homogenizing B1+ in cardiac imaging at 7T.

Stefanie Buchenau¹, Martin Haas¹, Juergen Hennig¹, and Maxim Zaitsev^1
1Department of Radiology, Medical Physics, University Medical Center Freiburg, Freiburg, Germany

Published methods for estimating specific absorption rate via post-processing of measured transmit fields are restricted to transmission and receive configurations in quadrature or RF shimming mode. Focusing on the challenge of estimating unknown magnetic field components within the present study an extension of the existing method is demonstrated to make it applicable to arbitrarily excited multi-transmit arrays. Based on simulations the quality of this extended method is investigated as a function of z position within a particular transmit array designed for parallel transmission experiments at 3T.

Ulrich Katscher¹, Christian Findeklee¹, and Tobias Voigt^1
1Philips Research Europe, Hamburg, Germany

The additional degrees of freedom in parallel transmission hamper straight-forward SAR estimations as applied for single channel transmit systems. Recently, a method was presented estimating SAR from B1 maps, which, however, was restricted to quadrature volume coils due to difficulties distinguishing phase contributions from RF transmission and reception. This study presents a method separating these two phase contributions, and thus, enables SAR estimation not only for quadrature volume coils, but also for single elements of a transmit array. The good, quantitative agreement between simulated and experimental phantom results found in this study underlines the feasibility of the proposed method.

Joonsung Lee¹, Matthias Gebhardt², Lawrence L Wald^3,4, and Elfar Adalsteinsson^1,4
1Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, ²Siemens Healthcare, Erlangen, Germany, ³A. A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States, ⁴Harvard-MIT Division of Health Sciences and, Massachusetts Institute of Technology, Cambridge, MA, United States

Parallel transmit (pTx) applications in MRI are limited by local SAR constraints. The peak local SAR can be estimated by monitoring so-called virtual observation points (VOPs) instead of searching exhaustively over all voxels in a 3D model. The VOPs can be pre-computed once for a given model and array configuration and applied in subsequent computation to efficiently estimate peak local SAR due to a given pTx RF pulse. We present a generalization of the original model compression method by VOPs, whereby we maintain the same accuracy in peak local SAR estimates, but with a reduced number of VOPs.

AbdELMonem M. El-Sharkawy¹, Di Qian^1,2, Paul A. Bottomley^1,2, and William A. Edelstein^1
1Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, Maryland, United States, ²Electrical and Computer Enginnering, Johns Hopkins University, Baltimore, Maryland, United States

Accurate and independent real-time measurement of RF power deposition is essential for MRI safety. We designed and built a six channel MR compatible, real-time power profiling system suitable for MRI up to 400MHz. The bench-calibrated system is highly linear over a 90dB dynamic range and does not interfere with scanner operation. RF power deposition was measured in eleven volunteers in a 3T scanner during whole-body MRI. The results showed considerable differences between the true power delivered and that estimated by the scanner supporting the view that scanner specific absorption rates do not accurately reflect true individual RF exposure.

Devashish Shrivastava¹, Ute Goerke¹, Shalom Michaeli¹, Jingeng Tian¹, Lance DelaBarre¹, and John T Vaughan^1
1University of Minnesota, Minneapolis, MN, United States

Total proton resonance frequency (PRF) shift coefficient was measured in the porcine brain to image radiofrequency heating with sub-degree C accuracy in ultra-high field MRI. The total PRF shift coefficient accounted for the effects of the change in molecular screening, volume susceptibility, and conductivity on the local magnetic field with temperature. The total PRF shift coefficient was measured as -0.02 ppm/0C and was significantly different from the traditionally used molecular shielding based PRF shift coefficient of -0.01 ppm/0C.