Variation of whole body averaged phantom specific absorption rate (SAR) in seven different 1.5 T MR systems

This paper evaluates the whole body averaged specific absorption rate (WB-SAR) for a saline-filled standardized phantom in seven different commercial 1.5 T MR systems. Using calorimetry we measured the actual delivered WB-SAR to the phantom. The variability between different MR systems of the actual WB-SAR delivered to the phantom strongly depends on the landmark positions; it ranges for a system estimated WB-SAR of 2 W/kg from 0.6 W/kg at landmark L1 and L2 to 1.2 W/kg for landmark L3. The WB-SAR overestimation ranges from 1.7 to 8.0 depending on the landmark, MR system, and temperature sensor used. The system estimated WB-SAR overestimates the phantom WB-SAR for most cases which can lead to a proportional underestimation of the reported implant heating. Testing in various MR systems using calorimetry to assess the WB-SAR is necessary because of large implant and gel heating variations in different MR systems.