Encrypted login | home

Program Information

On the Propagation of Proton Pencil Beams in Various Media


F Van den Heuvel

F Van den Heuvel*, F Fiorini , B George , University of Oxford, Oxford, Oxford

Presentations

SU-I-GPD-T-122 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall


Purpose: In modern proton therapy scanned pencil beams (PBs) are used to conform the dose to the target both at the distal and proximal ends of the volume. A homogeneous dose distribution is achieved by combing PBs of different energies, spot-sizes and positions. These properties are verified in water during commissioning of the beamline and the treatment planning system. However, the behavior of PBs varies in different materials. To model this, most planning systems use a water equivalent path length (WEPL) scaling. Here, we investigate how PBs change in different materials and densities, and quantify the suitability of the WEPL approach.

Methods: We use a Monte Carlo simulation model (FLUKA) to represent PBs from a clinically commissioned scanned-beam IBA facility. The PB dose distributions are calculated in water and compared to measured data. The simulations are repeated in water with different densities (0.1 to 1 g/cm3), muscle, adipose tissue, bone, and lung. For all simulations the dose planes perpendicular to the beam axis are parameterized using a stable distribution of the form Se(z; α, γ), where α quantifies the tails of the distribution and γ the broadening. The changes in the parameters α and γ over all materials are investigated.

Results: Both α(z) and γ(z) vary depending on density and material. Using WEPL, α(z) can be matched to those in water and for materials and effective WEPL factor can be found when densities do not differ greatly (0.7 to 1 g/cm³). WEPL is not effective for γ(z), but a geometric correction can be found for a density, ρ with geometric parameters a and b: γ(z/ρ) - a zᵇ maps to the variation of γ in water. Here: a=0.015,b=2.33

Conclusion: We have shown that PBs vary depending on material and density not corrected using WEPL scaling. A suitable correction is presented.


Contact Email: