A novel methodology for the optimization of transmission and dosimetric leaf gap parameters.

PURPOSE: Optimization of dosimetric leaf gap (DLG) and transmission is commonly performed through a manual trial and error process, which can lead to sub-optimal values. The purpose of this work is to create an alternative automated optimization process that provides the optimal DLG and transmission pair for use in a clinical setting. METHODS: Utilizing the treatment planning system application programming interface, a phase space of clinically viable DLG and transmission pairs was generated. The phase space contained 51,051 dose planes for DLGs between 0.0  and 2.5 mm and transmission values between 0.01% and 2.5%. Thirteen plans were measured for multiple multileaf collimator types and nominal beam energies. The optimization minimized the mean γ-index and maximized the γ-index pass rate. The optimized values were validated using five plans excluded from the optimization. RESULTS: Of the nominal beam energies and multileaf collimator system (MLC)-type combinations tested, 6/7 showed an increase in γ-index pass rate and a decrease in mean γ-index signifying better agreement between measurement and calculation. When comparing the optimized DLG and transmission values to the clinically implemented values identified via an iterative method, 5/7 energy, and MLC type combinations showed no statistically significant changes. In addition, the optimized values were benchmarked against three Task Group 119 plans with published γ-index pass rates, which had been held out of the optimization. For those plans, the optimized DLG and transmission values provided the same or better γ-index pass rates. CONCLUSION: We presented a novel and viable automated alternative to current approaches of selecting the DLG and transmission parameters. This method will reduce the time required to determine the clinically acceptable DLG and transmission parameters and ensure optimality for the plans included in the optimization.