SU-E-T-716: An Investigation of Heterogeneity Dose Calculation Accuracy in IMRT from Five Commercial Treatment Planning Systems Useing Monte Carlo

Abstract

Purpose: The goal of this study was to compare the accuracy of dose calculations relative to Monte Carlo calculations within a low‐density region of five common IMRT TPSs. Methods: Ten lungcancerIMRT cases were optimized with five different TPSs. Of the five TPSs we tested, Three use CSA/AAAs(Xio4.0, Pinnacle8.0m and Eclipse7.5.18.0) and two use PBAs(Corvus08, Eclipse7.5.18.0) dose calculation algorithms. All plans were designed for delivery on a Varian 23EX accelerator using 6MV photon beam. All plans were recalculated by Monte Carlo using leaf sequences and MUs for individual plans. Dose calculation algorithms were evaluated based on the criteria by Task Group 53. Profile doses, 2D dose distributions, point dose and Gamma index were compared. Results: The dose predicted in the target center met the criteria (±5% or ±3mm) for all TPSs compared to Monte Carlo calculated results; However, there were greater variations (from3.2% to10.7%) in the low‐dose region. The CSA/AAAs performed good agreement within the surrounding lung regions, even in the penumbral region compared with Monte Carlo methods. The CSA/AAAs greater than 95% pixels met the criterion at the ±7%/±7mm agreement level and more than 87% of the pixels still met even when the criterion was tightened to ±5%/±3mm. However, the agreement for the PBAs fewer than 66% pixels met the ±7% /±7mm criterion, with only 52% of the pixels meeting the ±5% /±3mm criteria. Conclusions: The Monte Carlo dose calculation system provides a platform for the fair comparison and evaluation of treatment plans to facilitate clinical decision making in selecting a TPS and beam delivery system for particular treatment sites. Based on our analysis, CSA/AAAs performs consistently and more accurately than PBAs when applied to IMRT planning involving low‐density regions.