Uncertainty of scintillator-based field-output factor measurements in MR-Linacs with the two-channel chromatic stem removal technique

Abstract

Current international protocols such as the IAEA/AAPM TRS-483 for small field dosimetry do not include recommendations for measurements in strong magnetic fields as found in MR-Linacs (i.e. megavoltage x-ray radiotherapy units with built-in magnetic resonance imaging). A EURAMET research project was therefore launched to support the traceability of such measurements and to contribute to the metrological basis for future recommendations. Given the high degree of water equivalence and the small size of fiber-coupled plastic scintillators, one objective of the project was to study the suitability of this detector type for dosimetry in MR-Linacs and to demonstrate their use in a clinical environment. This manuscript reports results of that work. A protocol for measuring field output factors with plastic scintillation detectors was developed. The protocol is based on calibration of the scintillators directly in a water phantom positioned in the MR-Linac rather than in a water-equivalent solid phantom as has been a common technique for conventional linacs. A detailed uncertainty budget was established involving all known sources of uncertainty including the systematic influence of light attenuation through the fiber when measuring in different field sizes. The protocol was used to measure field output factors at five MR-Linac’s (four Viewray MRIdian’s and one Elekta Unity). The protocol showed a high level of robustness leading to a less than 0.3 % deviation between field output factors calculated based on calibration data obtained before or after the field-output factor measurements. The correction for light attenuation was less than 0.4 % for all fields measured but showed a systematic offset, with larger corrections needed for small field sizes.This robustness of the protocol was further demonstrated by the closeness of agreement (agreement within uncertainty) for field-output factor measurements with two different scintillators (BCF-12 and BCF-60) for square fields (size: 0.83 cm x 0.83 cm or larger) in four nominally identical Viewray MRIdian's. To test the accuracy of the scintillator measurements, the system was used for characterization of a PTW 60019 micro diamond detector in a Viewray MRIdian. Scintillator based correction factors for output factor measurements with the micro diamond detector were found to be in agreement with independent Monte-Carlo computations. The study therefore supports that plastic scintillation detectors are an excellent option for measuring field output factors in MR-Linacs.