Advances in Monte Carlo simulations for the design of X-ray phase contrast imaging systems

X-ray phase contrast imaging is known for its potential to yield high contrast in soft tissue and light materials compared to conventional transmission contrast, making it a valuable tool for (bio)medical applications and non-destructive testing. In addition to transmission and phase contrast, dark field contrast is a third contrast type that has received increased interest recently due to its unique ability to detect the presence of unresolved microstructures. Phase and dark field contrast are both related to the material’s X-ray refractive index, but as opposed to phase contrast, dark field contrast results from refractive index fluctuations that cannot be resolved by the imaging system. Remarkably, all three contrast types can be measured in a single experiment using dedicated X-ray phase contrast imaging systems. The introduction of compact lab-based systems for phase sensitive X-ray imaging has resulted in a growing number of applications. Optimization of conventional X-ray imaging setups often relies on Monte Carlo simulations, where stochastic models are used to simulate the physics processes. In this work, a collection of tools is presented that aim to meet the requirements associated with Monte Carlo simulations for the design of X-ray phase contrast imaging systems. Here, two X-ray phase contrast imaging methods are mainly of interest: grating-based interferometry and edge illumination, each of which relies on the use of gratings. First, the realization of X-ray phase contrast simulations for grating-based interferometry is demonstrated using the GATE Monte Carlo framework, hereby relying on a hybrid simulation approach combining Monte Carlo simulations with wave optics calculations. Although this simulation framework is suitable for edge illumination simulations as well, Monte Carlo simulations are known to be very time consuming, certainly for parameter studies. To address this limitation, the concept of virtual gratings is introduced. By replacing the gratings in the simulation with virtual gratings, the parameters of the gratings can be changed after the simulation, thereby significantly reducing the overall simulation time. This concept is subsequently used for the design of edge illumination gratings for the augmentation of the FleXCT micro-CT scanner to a phase sensitive X-ray imaging system. Finally, the benchmarking of multi-contrast X-ray imaging simulations is addressed. The reference values required for benchmarking are particularly difficult to determine for the dark field contrast. Here, a practical method based on the virtual grating approach is presented to directly estimate reference values for all three contrast types from the simulated X-ray trajectories, allowing for efficient benchmarking.