Program Information
A Test Suite for Optimization and Verification of MLC Treatment Planning System Modeling
R Popple1*, M Barnes2 , E Covington2 , K Farrey3 , J Moran4 , M Perez5 , D Raxter4 , T Ritter6 , (1) The University of Alabama at Birmingham, Birmingham, AL, (2) Calvary Mater Newcastle, Waratah, New South Wales, (3) University of Chicago, Chicago, IL, (4) University of Michigan Medical Center, Ann Arbor, MI, (5) Royal North Shore Hospital, St Leonards, New South Wales, (6) Richmond Veterans Administration Medical Center, Richmond, VA
Presentations
SU-I-GPD-T-603 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: To develop and apply a set of MLC patterns to comprehensively evaluate hardware variability and dose calculation algorithms.
Methods: Dynamic MLC control point sequences were created for constant velocity leaf motion and leaf separations 2-20 mm. Two sequences were developed for each separation, one for which all leaves moved together (UNIFORM) and one for which first odd leaves moved then even leaves (ALTERNATING). Static field-in-field sequences were comprised of alternating groups of open and closed leaves, with 1-8 leaves per group (SHUTTERS). The sequences were delivered with a 6 MV beam using four linear accelerators having 2.5mm leaf width MLCs at three institutions. Dose was measured using a thimble ionization chamber with length sufficient to span multiple leaves to ensure that inter- and intra-leaf leakage were averaged. The effective dynamic leaf gap (DLG) was calculated by extrapolating the UNIFORM measurements to zero gap. Average transmission was measured with leaves closed. Values were compared with treatment planning system calculations.
Results: Transmission values were similar between the four MLCs (1.14% to 1.22%); however, the DLG demonstrated significant variation (0.22 mm to 0.89 mm). The mean ratio of the measured to calculated value ranged from 0.97 to 1.0 for UNIFORM, and 1.02 to 1.05 for ALTERNATING. The agreement with calculation for UNIFORM was negatively correlated to ALTERNATING and was dependent on the DLG parameter in the TPS. The measured dose for SHUTTERS was within 1% for all MLCs. Calculated dose for STANDARD was independent of calculation grid size, whereas dose changed by up to 5% for ALTERNATING and 6% for SHUTTERS.
Conclusion: The measured values were consistent between machines, whereas the calculations were variable and depended on configuration and calculation parameters. This work can be used to define expected measured values for a given MLC and to guide improvements in beam modeling.
Funding Support, Disclosures, and Conflict of Interest: Dr. Moran acknowledges support from Varian.
Contact Email: