Monte Carlo Simulation for a Dynamic Bowtie Filter with Cone-beam CT

A pre-patient attenuator ("Bowtie filter" or "Bowtie") in X-ray computed tomography (CT) is used for reducing the radiation dose to patients and helping meet the detector dynamic range, and it also affects the CT image quality. Because a traditional static Bowtie filter is not optimal for modulating X-ray, we recently proposed a dynamic Bowtie filter for cone-beam CT, which could modulate X-ray more effectively. In this paper, we design and study the dynamic Bow-tie using Monte Carlo simulation. First, we establish a simulation frame, which contains a target phantom and a polychromatic source generated by a reflection target. Next, a design phantom is employed to help design the dynamic Bowtie for the target phantom. Finally, we obtain two phantom projections from two modes of filter:no Bowtie and the dynamic Bowtie. The results demonstrate that, under the action of the designed dynamic Bowtie, the X-ray dynamic range on detectors is highly compressed for the design phantom and the target phantom, which means that our methodology facilitate to uniform the photon flux. Moreover, with the uniformed photon flux, the radiation exposure suffered by tested objects reduces and the reconstructed image quality is improved.