Purpose: To determine a backlash of two general purpose multileaf collimators (MLCs) in the case of static and dynamic collimation.
Methods: To move a MLC leaf with a screw-type drive mechanism a certain amount of mechanical freedom, 'backlash', must be allowed for the screw to turn. In this work the backlash is described as a difference in an actual radiation beam width when a collimator is 'closed' to a drive position and when the collimator is 'opened' to the drive position. In the presence of a pure mechanical backlash: when the collimator is closed an actual beam width is larger than planned and when the collimator is opened an actual beam width is smaller than planned. The actual beam width of 'closed' and 'opened' collimation is measured with portal imager, average for the most center 10 cm of collimation is calculated and backlash projected at isocenter plane is estimated. Measurement is repeated for ten static drive positions and for dynamic drive sequence using three leaf speeds. Study is made with Varian Standard 120M (Varian Medical Systems Inc.) and Agility (Elekta Instrument AB) MLCs.
Results: A static backlash for studied MLC leafs is: 0.3 mm for Varian Standard 120M and 0.0 mm for Agility. The static backlash is found to be independent of the drive position. A dynamic backlash for studied collimator leafs is: 0.4 mm for Varian Standard 120M and -0.6 mm for Agility. In case of Varian Standard 120M the dynamic backlash is found to be independent and in case of Agility dependent of the leaf speed.
Conclusion: The backlash is found to limit the position accuracy of collimator leafs except in the case of a static treatment delivery with Agility MLC.
Funding Support, Disclosures, and Conflict of Interest: The first author holds a research physicist position financially supported by Varian Medical Systems Finland Ltd..
MLC, Beam Shaping, Collimation
Not Applicable / None Entered.