Optical diffraction tomography (ODT) has long been recognized for its potential in non-invasive, label-free imaging of live biological cells. However, a major challenge arises when ODT is applied to standard multi-well cell culture plates, a common tool use under realistic laboratory conditions for biological studies. The geometry of these cell culture plates restricts the range of oblique illumination angles, causing a loss of critical low-frequency information in the captured intensity images. This results in blurry details, reduced contrast, and degraded resolution in 3D tomographic reconstructions, especially when working in high-numerical-aperture (NA) systems. Despite the substantial potential of ODT for live-cell imaging, these challenges hinder its broad applicability for realistic biological environments. The difficulty of overcoming the illumination mismatch is particularly problematic for live-cell imaging in high-throughput settings, where multi-well cell culture plates are indispensable. This limitation prevents accurate tomographic reconstruction, which is necessary for analyzing fine cellular features, and thus impacts the overall effectiveness and efficiency of imaging systems in real-time biological monitoring.
To overcome these challenges, researchers from the Smart Computational Imaging Laboratory (SCILab) at Nanjing University of Science and Technology, led by Professor Chao Zuo, have developed a novel dark-field Fourier ptychographic diffraction tomography (DF-FPDT) technique. Their recent work was published in Volume 1, Issue 1 of iOptics on 19 August 2025. In this study, they introduce an innovative imaging framework that leverages non-matched illumination to enhance high-contrast live-cell imaging. By selectively updating high-frequency components inherent in intensity measurements and removing low-frequency background artifacts, DF-FPDT addresses the limitations of conventional optical diffraction tomography. This breakthrough technique enables high-resolution, high-contrast 3D reconstructions while retaining the volumetric, quantitative, and non-interferometric advantages of traditional Fourier ptychographic diffraction tomography (FPDT).
Prof. Zuo briefly explains the significance of this method, stating, “DF-FPDT uniquely leverages non-matched illumination to selectively update only high-frequency components, enhancing fine structural details without the need for additional hardware or post-processing. This provides a powerful tool for live-cell imaging, maintaining the core advantages of FPDT while addressing the limitations posed by standard laboratory setups.”
To validate the performance of DF-FPDT, both simulations and experimental measurements demonstrated its ability to enhance structural contrast and improve image quality. The 3D reconstructed tomograms generated using DF-FPDT show a strong structural similarity between the simulated and experimental results, and highlighting superior contrast and finer details compared to those obtained using the traditional FPDT algorithm. In live-cell imaging, DF-FPDT successfully visualized intricate subcellular structures and captured dynamic cellular processes, such as mitochondrial fusion and fission, offering real-time, high-contrast, and high-resolution imaging. These findings underscore the potential of DF-FPDT to achieve dark-field-like contrast in live-cell imaging and facilitate dynamic cellular monitoring under standard laboratory conditions.
Prof. Zuo highlights the transformative potential of DF-FPDT, stating, “DF-FPDT offers significant potential for widespread application in realistic laboratory settings, including drug screening, cellular analysis, and dynamic subcellular monitoring, opening new possibilities in biomedical research.”
Looking ahead, Prof. Zuo and his team plan to further refine DF-FPDT by integrating adaptive illumination strategies and data-driven approaches such as deep learning-based reconstruction, which could improve real-time imaging and further enhance contrast. Additionally, future developments may include dual-mode imaging, enabling users to toggle between contrast-optimized and full RI reconstructions depending on the application, significantly broadening DF-FPDT’s use in both dynamic live-cell imaging and quantitative analysis.
“These future enhancements will reveal even greater potential for DF-FPDT in a wide range of biomedical applications, accelerating discoveries in live-cell imaging and beyond,” said Prof. Zuo.
Source:
Nanjing University of Science and Technology
Journal reference:
Ullah, H., et al. (2025). Intrinsic dark-field Fourier ptychographic diffraction tomography under non-matched illumination. iOptics. doi.org/10.1016/j.iopt.2025.100006