Program Information
Experimental and Monte Carlo Investigation of Dose Calculation Accuracy for Lung SBRT Planning
J Rosenfield*, N Manohar , X Yang , X Dong , E Elder , K Higgins , A Dhabaan , Emory University, Atlanta, GA
Presentations
SU-I-GPD-T-591 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: To assess the dosimetric accuracy of the Acuros XB (AXB) and AAA dose calculation algorithms for lung SBRT treatments through phantom measurements and Monte Carlo calculations.
Methods: To evaluate the accuracies of the algorithms in highly heterogeneous media, a simple treatment plan was delivered to a slab phantom with dose calculated using both AAA and AXB. The slab phantom consisted of solid water and a low-density insert representing lung. Field sizes of 3×3 cm² through 12×12 cm² were considered. The delivered dose was measured with OSLDs and radiochromic film and compared to the calculated dose distributions. Additionally, Monte Carlo calculations were performed using Varian VirtuaLinac for comparison with the measured slab phantom dose distributions. A VMAT lung SBRT plan was delivered to the thorax of an anthropomorphic phantom, and the dose was measured with radiochromic film. AAA and AXB calculated dose distributions were compared to the measured film dose using a 2D gamma analysis.
Results: The OSLD and film measurements for the slab phantom showed that AXB dose calculations are generally more accurate in heterogeneous media. The difference between the algorithms becomes more significant as field size decreases. Monte Carlo calculations predicted a decreased deposition of dose in the lung relative to water, which was observed with the film measurements, OSLD readings, and AXB calculations. The anthropomorphic phantom measurements were closer to the AXB calculations with respect to percentage of pixels passing the gamma analysis, though the difference between the AAA and AXB distributions was not dosimetrically significant.
Conclusion: Our results identify AXB as a more accurate dose calculation algorithm for lung SBRT planning. Further investigation with Monte Carlo calculations will provide additional insight regarding the clinical value of using AXB for constructing these plans.
Contact Email: