Two-argument total scatter factor for small fields simultaneously collimated by MLC and jaws: application to stereotactic radiosurgery and radiotherapy. (14th December 2017)

• Record Type: Journal Article
• Authors: Han, Zhaohui; Friesen, Scott; Hacker, Fred; Zygmanski, Piotr
• Is Part Of: Physics in medicine & biology. Volume 63:Number 1(2018:Jan.)
• Publication Date: 2017-12-14
• Subjects: total scatter factor -- output -- small field -- head scatter -- phantom scatter; Biophysics -- Periodicals; Medical physics -- Periodicals; 610.153
• DOI: 10.1088/1361-6560/aa9872
• Languages: English
• ISSNs: 0031-9155

Abstract

Abstract: Direct use of the total scatter factor ( S tot ) for independent monitor unit (MU) calculations can be a good alternative approach to the traditional separate treatment of head/collimator scatter ( S c ) and phantom scatter ( S p ), especially for stereotactic small fields under the simultaneous collimation of secondary jaws and tertiary multileaf collimators (MLC). We have carried out the measurement of S tot in water for field sizes down to 0.5 × 0.5 cm 2 on a Varian TrueBeam STx medical linear accelerator (linac) equipped with high definition MLCs. Both the jaw field size ( c ) and MLC field size ( s ) significantly impact the linac output factors, especially when c ≫ s and s is small (e.g. s < 5 cm). The combined influence of MLC and jaws gives rise to a two-argument dependence of the total scatter factor, S tot ( c, s ), which is difficult to functionally decouple. The ( c, s ) dependence can be conceived as a set of s -dependent functions ('branches') defined on domain [ s min, s max = c ] for a given jaw size of c . We have also developed a heuristic model of S tot to assist the clinical implementation of the measured S tot data for small field dosimetry. The model has two components: (i) empirical fit formula for the s -dependent branches and (ii) interpolation scheme between the branches. The interpolation scheme preserves the characteristic shape of the measured branches and effectively transforms the measured trapezoidal domain in ( c, s ) plane to a rectangular domain to facilitate easier two-dimensional interpolation to determine S tot for arbitrary ( c, s ) combinations. Both the empirical fit and interpolation showed good agreement with experimental validation data.