Modeling the Fluence Output Factor for TomoTherapy Dynamic Jaw Delivery

Abstract

Purpose/Objective(s): The dynamic jaw delivery improves the longitudinal resolution and throughput of the TomoTherapy treatment system. This technique utilizes small jaw widths near longitudinal borders of a tumor for higher resolution and utilizes large jaw widths in the middle for higher efficiency. Due to the finite spot size of the electron source, the fluence output factor (FOF) decreases as the jaws close, which results in the longitudinal borders of the target receiving fewer doses. To address this issue, we propose a model to calculate the FOF for different jaw settings. This model provides a quick solution to convert the fluence-equivalent jaw width to the light-projected jaw width. Materials/Methods: For TomoTherapys dose calculation, the integral fluence is more relevant than the center value of the jaw fluence profile. FOF in TomoTherapy is defined as the ratio of longitudinal integral fluence to the jaw width. The dynamic jaw algorithm generates jaw trajectories based on the assumption that the FOF remains the same for different jaw widths. However, the physical, light-projected jaw width which the initial MLC open sinogram is based on does not preserve FOF. The light-projected jaw width leads to a decrease of the FOF for small jaw opening. This could result in a sinogram with hot regions near the longitudinal borders for dynamic jaw delivery. To address the issue, we define the fluence-equivalent jaw width as the jaw width that preserves FOF and propose a fluence calculation model with a radiation source of finite size. A 2D Gaussian distribution is assumed for the photon source. Photons are assumed to be absorbed immediately once they are blocked by the jaws. The fluence value of a point on the isocenter plane is the integral of the part of the source that can be seen by the point. Photons in different directions have equal weight since the angular variation is very small within a few degrees in the forward direction. Longitudinal fluence profiles for different jaw settings are calculated based on the jaw geometry. Results: As the jaws close, source occlusion makes the FOF deviate from constant and eventually drop to zero. Different spot sizes were used to test the model. The results are plotted in the FOF vs. light-projected jaw width graphs. The results match well with the Monte Carlo simulations for different spot sizes. A spot size with a FWHM of 1.1 mm fits well with measurement results. The model has adequate accuracy to calculate the integral fluence. Conclusions: The variation of the FOF is mainly due to the finite size of the radiation source and its occlusion by the jaws. The simple model based on jaw geometry can calculate the FOF with adequate accuracy. Conversion from the fluence-equivalent jaw trajectory to light-projected jaw trajectory can be built based on this model to improve the efficiency of the dynamic jaw delivery.