Should We Care About Delivery Time in IMRT Planning?
K Mittauer*, B Lu, G Yan, D Kahler, C Liu, Univ Florida, Gainesville, FLSU-E-T-607 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To model the delivery time for step and shoot IMRT through developing a method for the user to characterize their machine. To investigate the impact of IMRT parameters on delivery time and recommend planning strategies in light of delivery time and plan quality.
Methods: A two-step delivery time model was derived, and verified against 33 delivered IMRT plans using an Elekta LINAC. Eleven head and neck patients with preexisting IMRT plans were selected for this retrospective study. The Pinnacle TPS was used to compute new plans by varying modulation parameters. 41 plans per patient (451 plans total) were generated with the following variations: 12 minimum segment area parameter (MSAP) plans, 9 minimum MU parameter (MMUP) plans, and 20 minimum MU and segment area parameter (MMUSAP) plans. Plans were evaluated based on DVHs (plan quality) and delivery time.
Results: The delivery time model accuracy was 1.8%, 10 s. Delivery efficiency improves above 5 MU for MMUP, but no improvement was found with MSAP. MMUP and MSAP had the greatest influence on the number of control points and the plan MU, respectively. MLC speed and gun delay were more costly than the plan MU. However, the plan MU could affect the delivery times for dose rate dependent plans. Threshold values for plan quality were 5 cm² for MSAP and 5 MU for MMUP.
Conclusion: The proposed formalism provides a valid method to assess delivery time and can be characterized for the users machine. For dose rate dependent cases or future deliveries with an enhanced MLC speed, treatment efficiency could be improved with MSAP and/or MMUP. But for most current clinical cases, only MMUP can improve delivery efficiency. We, however, do not recommend increasing these parameters for the sake of delivery efficiency at a cost to plan quality.