Room: Exhibit Hall | Forum 7
Purpose: Accurate dose calculation is essential for achieving the dosimetric advantage of intensity modulated proton therapy. A pencil beam(PB) algorithm, which use dual depth dose model to achieve high accuracy, was developed.
Methods: The algorithm uses dual depth dose-lateral spread model, K(r,z)=DDMCS(z)Ã—Î¦MCS(r,z) +DDNI(z)Ã—Î¦NI(r,z), which stipulates that multiple Coulomb scattering (MCS) and nuclear interaction protons have different depth dose and lateral profiles since they have distinctive properties. Depth dose of MCS protons was calculated using stopping power and range straggling, while the other depth dose is derived by subtracting the MCS part from total depth dose. Lateral profile of MCS protons was Gaussian distribution calculate by Fermi-Eyges theory, but for nuclear interaction lateral profile, another Cauchy distribution was used to better describe the lateral spread-out caused by nuclear interaction. To better treat lateral inhomogeneity, firstly a broad proton scanning beam is splitted into multiple thinner pencil beams hexagonally distributed around the original beam center, which is more efficient than conventional orthogonal grid division. Furthermore, during the transport of each pencil beam, the stopping/scattering effects were calculated with weight averaging all the materials within the pencil beam region, rather than just on the beam axis.
Results: The algorithm was benchmarked against Monte Carlo (MC) simulation with both homogeneous and heterogeneous phantoms to test its accuracy. Five beam energies, spanning the energy range with clinical interest, have been investigated. The maximum dose difference between the PB algorithm and MC result is about 1.8% for homogeneous cases, 2.3% for the heterogeneous cases, and planar dose gamma analysis yields pass rate >98% with 2%/2mm criteria and 10% dose threshold.
Conclusion: A PB algorithm based on dual depth dose-lateral profile was developed. The accuracy of the algorithm is tested and preliminarily verified, and further testing with clinical cases is underway.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by the National Natural Science Foundation of China (No. 81101132 and 11305203).