Comprehensive Study of Head and Neck IMRT Parameters On Planning and Delivery Efficiency, Plan Quality, and Dose Accuracy
K Mittauer*, B Lu, C Liu, G Yan, A Gopal, Univ Florida, Gainesville, FLSU-E-T-621 Sunday 3:00:00 PM - 6:00:00 PM Room: Exhibit Hall
To optimize planning and delivery efficiency, and quality of head and neck IMRT through the evaluation of planning parameters. This study also serves to identify the impact on dose accuracy due to calculation grid size.
Eleven head and neck patients, 45 trials per patient (495 trials in total), were evaluated varying IMRT parameters of dose grid, minimum MU per segment, minimum segment area, and control point number. Plans were recomputed on Pinnacle Treatment Planning System (TPS), and scaled to the planning target volume (PTV) constraint of 95% volume. Differential dose volume histograms (DVHs) were exported, and a program was written to compile DVH results. Plans were delivered on an Elekta Synergy linear accelerator to assess delivery time. Plan quality, calculation time, and delivery time served as this study's endpoints.
The 4 mm dose grid with 2 mm fluence grid in each direction, saving 1/3 the computation time, were most comparable by DVH results to the 2 mm dose and fluence grid. Dose uncertainty due to dose calculation grid effect was as high as 8.2%, 5.5 Gy for PTVs and 13.3%, 2.1 Gy for organs at risk. Smaller volumes and high gradient regions were more susceptible to uncertainties. Threshold values that maintained adequate plan quality were 5 cm² for minimum segment area and 5 MU for minimum MU. Minimum MU was more costly in terms of plan quality compared to the minimum segment area.
DVH differences can be effectively used to quantify the dose grid calculation uncertainty. For minimum MU and segment area, the DVH differences are an effect of the intensity map, defined by MLC shape and the number of control points. Exceeding the adequate number of control points diminishes returns of plan quality and increases patient treatment time.