A new convolution/superposition dose calculation method for external photon beam radiation therapy using beam modifiers and/or independent collimators

Abstract

This thesis was dedicated to developing a convolution/superposition dose calculation method for computing patient dose distributions as well as radiation output (monitor units) for photon radiotherapy. A new dual source photon beam model based on Monte Carlo simulation of clinical linear accelerators was developed and incorporated in the convolution method. This allowed for more accurate computation of dose distributions and output factors for photon fields, because both the primary and extra‐focal photons from clinical accelerators can be accounted for accurately. This convolution method was tested extensively for a variety of photon fields including those using independent jaws and/or beam modifiers, such as wedges and blocks. To use this convolution method for wedged fields, an extended phantom model was developed to integrate a wedge along with a patient phantom. Thus, dosimetric effect of wedges regarding beam hardening effect and secondary radiation generated by the wedge was taken into account. The new convolution method was also applied to calculate dose and output factors for dynamic or intensity modulated fields such as dynamic wedges. To use the convolution method for clinical situations, corrections for beam divergence and beam hardening were also developed. The parameters for the convolution calculation related to spatial resolution were investigated to optimize the speed and accuracy of the convolution calculation.