Adaptive-learning physics-assisted light-field microscopy enables day-long and millisecond-scale super-resolution imaging of 3D subcellular dynamics.

ABSTRACT: Long-term and high-spatiotemporal-resolution 3D imaging of living cells remains an unmet challenge for super-resolution microscopy, owing to the noticeable phototoxicity and limited scanning speed. While emerging light-field microscopy can mitigate this issue through three-dimensionally capturing biological dynamics with merely single snapshot, it suffers from suboptimal resolution insufficient for resolving subcellular structures. Here we propose an Adaptive Learning PHysics-Assisted Light-Field Microscopy (Alpha-LFM) with a physics-assisted deep learning framework and adaptive-tuning strategies capable of light-field reconstruction of diverse subcellular dynamics. Alpha-LFM delivers sub-diffraction-limit spatial resolution (up to ~120 nm) while maintaining high temporal resolution and lo

SUBMITTER: Zhu L 

PROVIDER: S-EPMC12322278 | biostudies-literature | 2025 Aug

REPOSITORIES: biostudies-literature