Program Information
A Sensitivity Analysis for Evaluating Dosimetric Impact of MLC Modeling Parameter Accuracy On IMRT Treatment Plans
S Utsunomiya1*, N Shibuya2 , M Yamaki2 , R Sasamoto2 , H Aoyama1 , (1) Graduate School of Medical and Dental Sciences, Niigata University, Niigata, (2) Graduate School of Health Sciences, Niigata University, Niigata
Presentations
SU-E-T-392 Sunday 3:00PM - 6:00PM Room: Exhibit HallPurpose: To quantify and evaluate the dosimetric impact of MLC modeling parameter accuracy on IMRT treatment plans by performing a sensitivity analysis.
Methods: Physician-approved 8 prostate and 4 head-and-neck IMRT treatment plans in which 7 and 9 fields dynamic MLC technique was utilized respectively were retrospectively analyzed. The dose re-calculation was performed with Eclipse treatment planning system (ver.8.9, Varian Medical Systems) after MLC modeling parameters such as leaf transmission factor (LTF) and dosimetric leaf gap (DLG) were manually shifted from the original value. The change in DVH parameters (PTV D95%, rectum wall D25% and bladder wall D35% for prostate cases; PTV D95%, parotid gland mean dose and spinal canal maximum dose for head-and-neck cases) and point doses were plotted as a function of errors in MLC modeling parameters and the dosimetric impact was quantified by sensitivity (slope of an approximate linear function). The 2D map of dose sensitivity to LTF and DLG were created and overlaid onto CT image of the patient using MATLAB software (ver.8.2.0.701, MathWorks).
Results: The changes in all analyzed DVH parameters had almost linear relationship with errors in LTF and DLG. The sensitivities of DVH parameters for head-and-neck cases were higher than prostate cases overall. For prostate cases, sensitivities of DVH parameters to DLG were higher than LTF and the opposite results were obtained for head-and-neck cases. According to the 2D map, LTF and DLG had high dose sensitivity around PTV edge especially at the overlap between PTV and rectum for a prostate case. The region of high dose sensitivity to LTF was more blurred and widely scattered than DLG.
Conclusion: The dosimetric impact of MLC modeling parameter accuracy on IMRT treatment plans were successfully quantified by a sensitivity analysis of DVH parameters and a 2D map of dose sensitivity overlaid onto CT image of patients.
Contact Email: