An evaluation of quality assurance guidelines comparing the American College of Radiology and American Association of Physicists in Medicine task group 284 for magnetic resonance simulation

PURPOSE: The purpose of this study was to evaluate similarities and differences in quality assurance (QA) guidelines for a conventional diagnostic magnetic resonance (MR) system and a MR simulator (MR‐SIM) system used for radiotherapy. METHODS: In this study, we compared QA testing guidelines from the American College of Radiology (ACR) MR Quality Control (MR QC) Manual to the QA section of the American Association of Physicists in Medicine (AAPM) Task Group 284 report (TG‐284). Differences and similarities were identified in testing scope, frequency, and tolerances. QA testing results from an ACR accredited clinical diagnostic MR system following ACR MR QC instructions were then evaluated using TG‐284 tolerances. RESULTS: Five tests from the ACR MR QC Manual were not included in TG‐284. Five new tests were identified for MR‐SIM systems in TG‐284 and pertained exclusively to the external laser positioning system of MR‐SIM systems. “Low‐contrast object detectability” (LCD), “table motion smoothness and accuracy,” “transmitter gain,” and “geometric accuracy” tests differed between the two QA guides. Tighter tolerances were required in TG‐284 for “table motion smoothness and accuracy” and “low contrast object detectability.” “Transmitter gain” tolerance was dependent on initial baseline measurements, and TG‐284 required that geometric accuracy be tested over a larger field of view than the ACR testing method. All tests from the ACR MR QC Manual for a conventional MR system passed ACR tolerances. The T2‐weighted image acquired with ACR sequences failed the 40‐spoke requirement from TG‐284, transmitter gain was at the 5% tolerance of TG‐284, and geometric accuracy could not be evaluated because of required equipment differences. Table motion passed both TG‐284 and ACR required tolerances. CONCLUSION: Our study evaluated QA guidelines for an MR‐SIM and demonstrated the additional QA requirements of a clinical diagnostic MR system to be used as an MR‐SIM in radiotherapy as recommended by TG‐284.