Abstract
Recent advances in deep learning-based image reconstruction techniques have led to significant progress in phase retrieval using digital in-line holographic microscopy (DIHM). However, existing phase retrieval methods have technical limitations in 3D morphology reconstruction from single-shot holograms of biological cells. In this study, we propose a deep learning model, named MorpHoloNet, for single-shot reconstruction of 3D morphology by integrating physics-driven and coordinate-based neural networks. By simulating optical diffraction of coherent light through a 3D phase shift distribution, MorpHoloNet is optimized by minimizing the loss between simulated and input holograms on the detector plane. MorpHoloNet enables direct reconstruction of 3D complex light field and 3D morphology of a test sample from its single-shot hologram without requiring multiple phase-shifted holograms or angular scanning. It would be utilized to reconstruct spatiotemporal variations in 3D translational and rotational behaviors, as well as morphological deformations of biological cells from consecutive single-shot holograms captured using DIHM.