Abstract:

Radiotherapy outcomes are highly dependent on entire dose distributions. Patient inhomogeneities such as lung continue to pose a serious challenge to dosimetric accuracy of radiotherapy. Intensity Modulated Radiotherapy (IMRT) is an advanced form of radiotherapy that uses complex beam patterns to produce optimal dose distributions. However, the effects of patient inhomogeneities on these complex, intensity-modulated fields have yet to be understood in clinical terms. The irregular, small segment sizes, steep dose gradients, and dosimetric modeling challenges inherent to IMRT may be increasingly vulnerable to perturbing effects of patient inhomogeneities. An MCNP-based Monte Carlo virtual source model of a Novalis? radiotherapy treatment unit has been developed and validated. This source model is capable of simulating full conformal and intensity modulated radiotherapy treatment plans. As part of the source model, a novel method to efficiently model the micro Multileaf Collimator (mMLC) has been developed. This method represents and preserves detailed geometric characteristics of patient-specific mMLC configurations while accounting for spectral effects of photon transport. The accuracy of this source model has been validated using recommended criteria of the American Association of Physicists in Medicine. Additionally, an interface has been developed to integrate and optimize CT-based patient modeling with the virtual source model. The accuracy of the virtual source model and patient modeling interface have been validated with clinical IMRT Quality Assurance procedures. The virtual source model has been used to simulate and characterize the effects of low-density lung on Novalis? conformal and intensity-modulated radiotherapy beams. This Monte Carlo model, along with CT-based patient modeling interface, provide powerful dosimetric tools for additional clinical investigations and research.