SU‐E‐T‐427: Feasibility Study for Evaluation of IMRT Dose Distribution Using Geant4‐Based Automated Algorithms

Purpose: For intensity‐modulated radiation therapy (IMRT) treatment planning validation using Monte Carlo (MC) simulations, a precise and automated procedure is necessary to evaluate the patient dose distribution. The aim of this study is to develop an automated algorithm for IMRT simulations using DICOM files and to evaluate the patient dose based on 4D simulation using the Geant4 MC toolkit. Methods: The head of a clinical linac (Varian Clinac 2300 IX) was modeled in Geant4 along with particular components such as the flattening filter and the multi‐leaf collimator (MLC). Patient information and the position of the MLC were imported from the DICOM‐RT interface. For each position of the MLC, a step‐ and‐shoot technique was adopted. PDDs and lateral profiles were simulated in a water phantom (50×50×40 cm 3 ) and compared to measurement data. We used a lung phantom and MC‐dose calculations were compared to the clinical treatment planning used at the Seoul National University Hospital. Results: In order to reproduce the measurement data, we tuned three free parameters: mean and standard deviation of the primary electron beam energy and the beam spot size. These parameters for 6 MV were found to be 5.6 MeV, 0.2378 MeV and 1 mm FWHM respectively. The average dose difference between measurements and simulations was less than 2% for PDDs and radial profiles. The lung phantom study showed fairly good agreement between MC and planning dose despite some unavoidable statistical fluctuation. Conclusion: The current feasibility study using the lung phantom shows the potential for IMRT dose validation using 4D MC simulations using Geant4 tool kits. This research was supported by Korea Institute of Nuclear safety and Development of Measurement Standards for Medical Radiation funded by Korea research Institute of Standards and Science. (KRISS‐2015‐15011032)