The implementation of conventional ODT typically relies on vibration-sensitive laser interferometric measurements, which are prone to speckle artifacts and can degrade spatial resolution. Although non-interferometric ODT approaches based on LED illumination simplify system configuration and improve imaging quality, they still fundamentally depend on sequential illumination angle scanning to achieve sufficient Fourier spectral coverage. Consequently, imaging speed remains a major bottleneck, particularly for capturing rapid dynamic processes in live cells, such as membrane blebbing, cytoskeletal fluctuations, lipid droplet transport, and mitochondrial fission. Furthermore, extending ODT to partially coherent illumination generally requires mechanical axial scanning to acquire defocused intensity stacks, which imposes additional constraints on temporal resolution. Collectively, these limitations hinder the application of ODT to millisecond-scale subcellular dynamics and make it challenging to meet the demands of high-quality, long-term live-cell imaging.
To overcome these challenges, researchers from the Smart Computational Imaging Laboratory (SCILab) at Nanjing University of Science and Technology, led by Professor Chao Zuo, in collaboration with Professor Dayong Jin from the University of Technology Sydney, have developed a spatiotemporal-multiplexed Fourier ptychographic diffraction tomography system (STM-FPDT) . Their recent work was published in Volume 8, Issue 2 of Advanced Photonics on 15 Feb 2026 . In this study, the team proposed a novel non-interferometric ODT framework. By employing an NA-matched annular coherent/partially coherent (C/PC) hybrid illumination strategy, the method enables parallel acquisition of multi-angle information, allowing STM-FPDT to capture the same spectral information as conventional FPDT while reducing acquisition time by 4.5-fold. In addition, the spatiotemporal collaborative tomographic reconstruction framework, integrating nonlinear global optimization with a sliding window protocol and leveraging the spatiotemporal continuity prior of biological structures, effectively suppresses motion artifacts, enabling high spatiotemporal resolution observation and reconstruction of subcellular dynamic processes.
Elaborating on the findings, Prof. Zuo says, “STM-FPDT overcomes the fundamental speed limitations of conventional optical diffraction tomography by introducing a spatiotemporal-multiplexed imaging strategy based on hybrid coherent/partially coherent illumination. By leveraging the inherent spatiotemporal continuity of biological structures, the approach employs a spatiotemporal tomographic reconstruction framework and enables high-resolution, high-speed 3D imaging of dynamic cellular processes.”
Using their self-developed STM-FPDT imaging system, the research team conducted a series of experiments to evaluate its performance. Compared to conventional optical diffraction tomography, STM-FPDT significantly improved temporal resolution while maintaining high spatial resolution, achieving 347 nm lateral and 1.54 μm axial resolution with a 3D volumetric imaging rate of 5 Hz. During live-cell dynamic observations, including membrane blebbing, cytoskeletal fiber fluctuations, lipid droplet motion, and mitochondrial fission, the system enabled clear sub-second visualization of organelle dynamics while effectively mitigating motion-blur artifacts
They also outlined several future directions for STM-FPDT. One promising avenue is the integration of physics-based neural networks with hybrid model- and data-driven strategies, enabling more efficient utilization of spatiotemporal correlations in biological structures and enhanced 3D reconstruction under extreme spatiotemporal bandwidth constraints. Another key focus is multi-modal imaging, combining fluorescence molecular specificity or intrinsic anisotropy of samples with noninvasive diffraction tomography to provide richer and deeper optical information for studying complex biological processes in live cells.
“These advances may pave the way for the next generation of high-speed, high-resolution 3D cellular imaging,” the research team noted, “enabling new insights into subcellular dynamics and broadening applications in biological research and beyond.”
Advanced Photonics
Imaging analysis
Spatiotemporal-multiplexed Fourier ptychographic diffraction tomography for high-speed, label-free 3D imaging of live cells
15-Feb-2026
The authors declare that they have no competing interests.