Author: George X. Ding, Kenneth L. Homann 👨🔬
Affiliation: Vanderbilt University Medical Center 🌍
Purpose: It is standard practice that the beam monitor units (MUs) calculated by a commercial treatment planning system (TPS) is verified by a secondary dose calculation (SDC) software that independently checks TPS calculated beam MUs. For electron beams, the MU calculated by the SCD is calculated from output/cutout factors based on measurement. When a large discrepancy occurs, one must decide which MUs are more accurate for patient treatment. This study aims to identify the source of the discrepancy to determine the accurate MUs to use.
Methods: Discrepancies often occur when the patient treated body surface displays substantial curvature instead of a flat surface. Using the Monte Carlo simulation (BEAMnrc/DOSXYZnrc) and Varian Eclipse eMC system, we compared a 6 MeV electron beam incident on a flat surface relative to a curved surface where the surface curvature displays a 1 cm dip vertically in saddle curvature with a 4 cm horizontal width.
Results: Eclipse calculates 102 MU and 91 MU, respectively, on a flat and 1 cm dip saddle curvature surface to achieve 100 cGy maximum dose for a 6 MeV beam (10x10 field with SSD=101 cm). The MUs calculated by the SDC is 102 MUs for both cases since the calculations are independent of surface curvature and depend only on beam field size and SSD. A 12% discrepancy was seen between Eclipse and the SCD calculated MUs for the 1 cm dip saddle curvature case. Our Monte Carlo simulation calculates the same MU for both setups as Eclipse, highlighting the improved accuracy of MUs by Eclipse.
Conclusion: The current commercial TPS can provide accurate dose distribution and beam monitor units even when the surface is not flat while the SDC software cannot account the effect of patient body surface curvature variations. This is the root cause of large discrepancies.