Dosimetric validation of Acuros^® XB with Monte Carlo methods for photon dose calculations

TLDR

The study investigates the dosimetric accuracy of the Acuros® XB advanced dose calculation algorithm for single radiation fields in homogeneous and heterogeneous geometries, comparing it to the analytical anisotropic algorithm (AAA). The authors validated Acuros® XB by comparing ion‑chamber measurements of 6‑ and 18‑MV beams across field sizes in a unit‑density phantom and by benchmarking its performance against BEAMnrc/DOSXYZnrc calculations in lung, low‑density lung, air, and bone geometries, with AAA results also reported for comparison. Acuros® XB agreed with measurements and Monte‑Carlo benchmarks within ±1.9% in homogeneous fields, ±2.0%/±2.9% in lung and low‑density lung, ±1.5%–±4.5% around air cavities, and within ±2%/2 mm on a lung CT dataset, whereas AAA showed discrepancies up to 10.2%/17.5%, demonstrating that Acuros® XB achieves Monte‑Carlo‑level dose accuracy.

Abstract

Purpose: The dosimetric accuracy of the recently released Acuros ® XB advanced dose calculation algorithm (Varian Medical Systems, Palo Alto, CA) is investigated for single radiation fields incident on homogeneous and heterogeneous geometries, and a comparison is made to the analytical anisotropic algorithm (AAA). Methods: Ion chamber measurements for the 6 and 18 MV beams within a range of field sizes (from to ) are used to validate Acuros ® XB dose calculations within a unit density phantom. The dosimetric accuracy of Acuros ® XB in the presence of lung, low‐density lung, air, and bone is determined using BEAMnrc/DOSXYZnrc calculations as a benchmark. Calculations using the AAA are included for reference to a current superposition/convolution standard. Results: Basic open field tests in a homogeneous phantom reveal an Acuros ® XB agreement with measurement to within ±1.9% in the inner field region for all field sizes and energies. Calculations on a heterogeneous interface phantom were found to agree with Monte Carlo calculations to within ±2.0% in lung and within ±2.9% in low‐density lung . In comparison, differences of up to 10.2% and 17.5% in lung and low‐density lung were observed in the equivalent AAA calculations. Acuros ® XB dose calculations performed on a phantom containing an air cavity were found to be within the range of ±1.5% to ±4.5% of the BEAMnrc/DOSXYZnrc calculated benchmark in the tissue above and below the air cavity. A comparison of Acuros ® XB dose calculations performed on a lung CT dataset with a BEAMnrc/DOSXYZnrc benchmark shows agreement within ±2%/2mm and indicates that the remaining differences are primarily a result of differences in physical material assignments within a CT dataset. Conclusions: By considering the fundamental particle interactions in matter based on theoretical interaction cross sections, the Acuros ® XB algorithm is capable of modeling radiotherapy dose deposition with accuracy only previously achievable with Monte Carlo techniques.