TH-D-201B-03: A Comparison of Protons versus Gamma X-Rays in Producing Patient Radiographs: A Monte Carlo Study

Abstract

Purpose: To assess the imaging properties of protonradiography against X‐ray radiography for both a phantom and an actual patient lungcancer case using Monte Carlo simulations.Methods and Materials: The GEANT4 Monte Carlo toolkit was used to simulate radiographs with a pure 230‐MeV proton beam as well as a pure 100‐keV X‐ray beam both with similar dose equivalent. For those simulations a phantom was designed using a water cube in which different tissues (low density lung high density lung adipose soft tissuemuscle cartilage bone) were implemented each within 10 cylinders of different sizes. The simulations were run for different phantom and tissue thicknesses. For each material spatial resolution signal‐to‐noise ratio (SNR) and contrast‐to‐noise ratio (CNR) were compared for the two modalities. Using an actual CT a protonradiograph of a lungcancer patient was then simulated and compared to the actual portal X‐ray image. The contrast of soft‐to‐soft tissue (S2S) soft‐to‐bone tissue (S2B) and bone‐to‐bone (B2B) were evaluated for both radiographs.Results: X‐ray radiographs present a 1‐mm spatial resolution;protonradiographs show a larger 3‐mm spatial resolution due to multiple coulomb scattering and halo effect of the proton beam. SNR and CNR of a radiograph using X‐rays stay constant versus tissue area size (20dB). For a radiograph using protons both SNR and CNR increase from 2dB to 30dB. Protonradiographs present much better S2S and B2B contrasts but poorer S2B contrasts than X‐ray radiographs.Conclusion: Preliminary results show that protonradiography presents lower spatial resolution than X‐ray radiography. Conversely protonradiography is more sensitive to tissue density variations it offers better soft‐to‐soft and bone‐to‐bone tissue contrasts hence allowing for sharper edge detection and better tumor localization for specific tumor cases such as lungcancer. Furthermore protonradiography would significantly reduce the dose to the patient.