Nobutaka Mukumoto MS, Presenter: Nothing to Disclose

Katsutomo Tsujii MSC, , Abstract Co-Author: Nothing to Disclose

Tokihiro Yamamoto PhD, Abstract Co-Author: Nothing to Disclose

Hideki Takegawa MSC, , Abstract Co-Author: Nothing to Disclose

Hodaka Numasaki PHD, , Abstract Co-Author: Nothing to Disclose

Teruki Teshima MD, Abstract Co-Author: Nothing to Disclose

Purpose/Objective(s): For Monte Carlo clinical linear accelerator simulations and patient dose calculations, it is important to determine phase space parameters of the incident electron beam accurately. These parameters, such as mean energy and radial intensity distribution, have traditionally been determined by matching the calculated dose distributions with the measured dose distributions through the process of trial and error. The drawbacks of this process are very time consuming and requiring a lot of Monte Carlo simulation experience and computational resources. A histogram-based multiple source model (MSM) that can be tuned to match the source characteristics from the accelerator is needed to overcome this limitation. In this study, we present a novel MSM based on the characterization of the photon source from a clinical linear accelerator. Materials/Methods: The benchmark phase-space data (PSD) of a 6 MV photon beam from Varian Clinac was created by using the EGSnrc/BEAMnrc MC code and analysed for their characteristics, such as planar fluences, angular distributions, and energy spectra. The novel MSM presented in this study took dependencies of correlations among these characteristics into account. The accuracy of the dose calculation using the EGSnrc/DOSXYZnrc MC code was tested for two MSMs with different dependencies of correlations. In addition, the novel MSM reconstructed the PSD from the different phase space parameters of the incident electron beam by tuning the histogram distributions. The statistical uncertainty of all the MC dose calculation was within 1%. Results: The novel MSM improved accuracy from 1.69% to 0.60%, with measured data in homogeneous water phantom, by retaining additional correlations, such as the off-axis locations at the last interaction point and the radiation angle. Energy spectra based on the last interaction component, such as target, primary collimator, and flattening filter, had the significant correlation with each particle not only the off-axis location at the PSD scoring plane, but also the additional correlations. The accuracy of the originally reconstructed PSD compared with that of the benchmark PSD in the inhomogeneous condition was 0.25%, although that of the base PSDs used for the reconstructed PSD was 1.41% and 0.74%, respectively. Conclusions: We developed the accurate MSM based on the characterization of the photon source from a clinical linear accelerator. The MC dose calculations of the novel MSM agreed with both those of the benchmark PSD and the measured data within 1%, for homogeneous and inhomogeneous conditions. The developed histogram-based MSM with the accurate characteristics of photon source has potential to facilitate the commissioning procedures by allowing tuning the histogram distributions.

Mukumoto, N, Tsujii, K, Yamamoto, T, Takegawa, H, Numasaki, H, Teshima, T, Novel Multiple Source Model from a Clinical Linear Accelerator Developed by Monte Carlo Photon Source Characterization.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8502410.html