Sperduto_Khan's Treatment Planning in Radiation Oncology, 5e

CHAPTER 20 Treatment Planning Algorithms: Photon Dose Calculations 449

●● Extrafocal radiation ●● Scatter and attenuation from beam-modifying devices ●● MC algorithms track histories of individual photons and electrons that undergo hard (“catastrophic”) collisions. Soft electron collisions are dealt with using condensed history methods. ●● The discrete ordinates method represents a numerical solution to the coupled Boltzmann transport equations. In this method, the energy, position, and direction of the radiation quanta are discretized for the numerical solution of the integro-differential equations. ●● Photon beam optimization presents additional challenges within the planning process. The need to calculate dose rapidly under conditions of changing incident fluence is necessary in a modern radiotherapy clinic. 5. The inelastic photon scattering processes that must be accounted for include A. Rayleigh scattering B. Moller scattering C. Photoelectric absorption D. Bremsstrahlung interactions 6. The convolution dose equation cannot be solved using Fourier analysis primarily because A. Scatter kernels are depth dependent B. Patient heterogeneities C. Beam hardening within the patient D. Step-size artifacts

●● Modern radiation therapy planning systems have evolved tremendously over the past few decades. A number of complex model-based photon dose algorithms exist which calculate dose to a 3D representation of the patient. These algorithms have been developed in response to improvements in algorithm development, computing power, and greater availability of volumetric imaging data. ●● Today, most commercial photon dose algorithms are a variation of the convolution/superposition method. As algorithm development and computing power improve, the use of MC and discrete-ordinates methods which better incorporate nonequilibrium dosimetry will likely increase. ●● A convolution/superposition model should account for the following characteristics: ●● Off-axis energy variations ●● Finite source size

REVIEW QUESTIONS

1. Which of the following algorithms is/are measurement based? A. Bentley–Milan B. Convolution/superposition C. Monte Carlo D. Discrete ordinates

2. Which of the following is/are used to speed up a Monte Carlo dose calculation? Choose all that apply. A. Variance reduction technique B. Condensed history method C. Kernel tilting D. Density scaling 3. Which of the following algorithms account for nonequilibrium conditions (e.g., at tissue interfaces)? Choose all that apply. A. Bentley–Milan B. Convolution/superposition C. Monte Carlo D. Discrete ordinates 4. For a typical 6-MV beam, an error in CT number of 2% leads to an error in dose of around A. 0.1%

Copyright © 2021 Wolters Kluwer, Inc. Unauthorized reproduction of the content is prohibited.

ANSWERS 1. A 2. A and B 3. C and D

4. B 5. C 6. B

B. 1% C. 5% D. 10%