Comparison of pencil beam and Monte Carlo calculations with ion chamber array measurements for patient-specific quality assurance

Objective To determine under what conditions and criteria comparisons between calculations made with the current clinical treatment planning system (Syngo) and an in-house built TPS (TIMPS) would allow skipping of in-beam portal-specific measurements. Methods Measurements were made with an array of 24 ion chambers in a water phantom for 227 proton and 313 carbon ion portals with and without a range shifter (RS). These measurements were compared with calculations performed with Syngo and TIMPS using metrics of average dose difference and Gamma index. Results For proton portals without RS, if a Gamma comparison between TIMPS and Syngo passed using criteria of 90% of tested points being within 3% and 3 ​mm, then 74% of measurements would agree with both TIMPS and Syngo. For proton portals with RS, more than 80% of measurements would agree with both calculations using the same criteria. For carbon ion portals without RS, if a Gamma evaluation between TIMPS and Syngo passed with criteria of 90% of tested points being within 2% and 2 ​mm, 85% of measurements would agree with both calculations. For carbon ion portals with RS, if a Gamma evaluation between TIMPS and Syngo passed with criteria of 90% of tested points being within 3% and 3 ​mm, 60% of measurements would agree with both calculations. Conclusions Both the pencil beam algorithm in Syngo and the FDC algorithm in TIMPS can provide accurate dose calculations in water for most clinical portals. For about 75% of portals, physicists can perform comparisons of calculations instead of phantom measurements to verify Syngo calculations thereby saving a large amount of beam time. There are some portals, however, such as for low-energy protons without RS and high-energy carbon ions, where agreement between the two calculations and measurements are not yet satisfactory to allow the elimination of all measurements.