SU‐E‐T‐878: Accurate Small MLC Field and Penumbra Modeling and Its Clinical Impact

Abstract

Purpose: To improve the accuracy of small MLC field and penumbra dose calculation by refining 6 MV beam model for IMRT‐based SBRT. Methods: A new dataset of depth and detector size corrected beam profiles of both jaw and MLC‐collimated fields were measured to refine a Pinnacle beam model for a Varian linac. Specific beam modeling sequence was used to tune the output factor for MLC‐collimated fields. In addition, the rounded leaf radius was also used as a free parameter for the new beam model. Previous IMRT patient QA data were also used for the verification of the new beam model and fine turning the MLC modeling parameters. The clinical impact of the refined model was studied by investigating the agreement of QA measurement with the new beam model. Results: Tests in MLC‐collimated fields showed our current beam model consistently over‐estimate the dose by as much as 4% for small MLC fields ranging between 2cm to 5cm with a large jaw opening. A refined new model with detector size corrected profiles required a smaller effective source size and a larger head scatter than the current beam model. We found the new model was able to reduce the IMRT QA discrepancies both in ion chamber and film measurements for SBRT of spine patients. Interestingly, our beam modeling process significantly improved the QA result for small field IMRT without affecting the QA result for large field IMRT cases. Conclusions: A more accurate Pinnacle beam model can be derived from the detector size corrected profiles, MLC‐collimated small field measurements, and the use of large amount of historical QA measurement data. The new beam model will benefit for intensity modulated spinal stereotactic radiosurgery or other small field IMRT applications.