Monte carlo estimation of dose in heterogeneous phantom around 6MV medical linear accelerator

Abstract

In this work, we completed a validation of the Varian Clinac IX equipped with the High Definition Multi-Leaf Collimator (HD 120 MLC) instead of the removable jaws, using GATE Monte Carlo Platform version 8.2. We validated the multileaf collimator (MLC) geometry by simulating two dosimetric functions (Percentage Depth Dose (PDD) and Dose Profile (DP)), for 6MV photon beam energy and different field sizes (3x3, 4x4, 6x6, 8x8, 10x10, 12x12, 15x15, and 20x20 cm2). We then compared the results with measurements realized with two detectors, namely the cylindrical ionization chamber and the micro-diode PTW silicon. By applying the Relative Dose Difference method (RDD), we noted a less than 2% and 1% agreement for the field sizes (10x10, 12x12, 15x15, 20x20 cm2) and (3x3, 4x4, 6x6, 8x8 cm2) respectively. Moreover, to evaluate the relevance of Monte Carlo method in a heterogeneous media, particularly in small field sizes (1x1, 2x2, 3x3 cm2), we have simulated three clinical studies based on the Physical Test Objects (PTOs) that are the equivalent slabs of lung and bone included in a water phantom. We noticed that the simulated PDDs exhibit two significant irregularities in the interface between water and lung. To eliminate these phenomena, we have used the "setMaxStepSizeInRegion" parameter implemented in GATE. We also noticed an important difference of 5% corresponding to the small field sizes, between homogeneous and heterogeneous simulated PDDs. We used the RDD method in this case as well. Moreover, we observed a difference between 1-4% between the simulated PDDs and the calculated ones by ECLIPSE Treatment Planning System (TPS). These results indicate that GATE (8.2) is useful in dosimetry with heterogeneous situations as well such as bone and lung.