Enhancing depth resolution in multislice x-ray ptychography using advanced reconstruction techniques

In this work, we explore novel strategies to enhance the depth resolution of multislice X-ray ptychography. Specifically, we investigate using sparse-angle ptychography tomography, employ virtual depth sectioning for refined reconstructions, and apply deep learning to mitigate crosstalk artifacts. These methods address complex scattering effects in thick specimens, offering a significant improvement over conventional single-slice models. We systematically evaluate their effectiveness using both simulated and experimental datasets, highlighting how they reduce feature blurring and random noise. Our findings pave the way for more accurate, high-resolution imaging at facilities like the upgraded Advanced Photon Source.