Program Information
Initial Experience Using Statistical Process Control Methods to Assess Mechanical Stability for a Prototype Straight-Through, Jawless, Fast-Gantry/fast-MLC Linac: Comparison to Truebeam
C Kennedy*, R Scheuermann , S Anamalayil , D Mihailidis , L Dong , J Metz , University of Pennsylvania, Philadelphia, PA
Presentations
TH-AB-FS1-9 (Thursday, August 3, 2017) 7:30 AM - 9:30 AM Room: Four Seasons 1
Purpose: We report our initial results using statistical process control methods to assess the mechanical stability of a prototype straight-through, jawless, fast-gantry/fast-MLC linac using data acquired from the Machine Performance Check (MPC) application. The prototype linac results were compared to results from three Truebeam linacs acquired over the same time period.
Methods: MPC results were acquired daily over 20 day period for the prototype linac and three Truebeam linacs. Two of the Truebeams were equipped with the Millennium 120 MLC, and one was equipped with an HDMLC. Process behavior limits were calculated using a subgroup size of one and three sigma limits for each mechanical component from the collected data. MLC results were calculated as the average and maximum values for each bank. The calculated upper thresholds for both the average and range charts were used as the metrics to compare mechanical stability between the machines.
Results: All linacs tested demonstrated excellent mechanical stability over the time period tested. The upper thresholds of the average charts relative to the average reading were typically less than 0.1mm, with worst-case values of 0.3 mm and 0.4 degrees for all mechanicals components. The upper thresholds of the range charts were less than 0.3 mm and 0.5 degrees for all components. The prototype machine compared favorably with the Truebeams, with process control limits generally within 0.1 mm/deg of the corresponding Truebeam component.
Conclusion: The prototype linac demonstrated excellent mechanical stability compared to the Truebeams we tested. These results represent a useful starting point for statistical process control analysis of Machine Performance Check mechanical results. This information could be used to detect drift in the mechanical component performance and determine the need for initialization/re-calibration or replacement. Additional data will be acquired to optimize the process behavior limits.
Funding Support, Disclosures, and Conflict of Interest: This work was performed as part of a research agreement with Varian Medical Systems.
Contact Email: