Plug-and-Play Volumetric Reconstruction for Compressive Sensing Light-Sheet Microscopy

2026-07-01

AI Summary

This work addresses the challenging high-aliasing three-dimensional reconstruction problem in compressed sensing light-sheet microscopy (CS-LSM) caused by multi-axis plane encoding. The authors propose a plug-and-play (PnP) optimization framework that innovatively incorporates an axial coupling model to capture inter-slice correlations. By integrating a fast data-consistency update based on the Woodbury identity with a Gauss–Seidel sweeping denoising strategy, the framework flexibly accommodates diverse user-defined denoisers. Under weakly convex regularization, the algorithm is guaranteed to have subsequence convergence. Experiments demonstrate that the method efficiently recovers fine cellular structures in both synthetic and real zebrafish heart datasets, significantly enhancing reconstruction continuity and quality, while also offering practical guidance for denoiser selection in CS-LSM applications.

Abstract

We investigate volumetric reconstruction for compressive sensing light-sheet microscopy (CS-LSM), where fast volumetric imaging is achieved by encoding multiple axial planes into each camera exposure. To recover the underlying volume from highly multiplexed measurements, we propose a plug-and-play (PnP) framework that flexibly incorporates any user-specified denoiser into the reconstruction process. Building on a slice-based formulation, we further introduce an axial-coupled model that exploits correlations between adjacent slices to improve volumetric continuity. For efficient computation, we derive a Woodbury-based update for the data-consistency step in both the slice-based and axial-coupled formulations, and employ a Gauss-Seidel sweep for the denoising step in the axial-coupled model. Under a weakly convex regularization assumption, we establish subsequential convergence of the proposed algorithm. Experiments on synthetic and real zebrafish-heart data demonstrate that the proposed framework successfully recovers cellular structures from compressed measurements, and provide practical insights into the comparative performance of commonly used denoisers within the PnP framework under the CS-LSM setup.

Problem

• volumetric reconstruction
• compressive sensing
• light-sheet microscopy
• multiplexed measurements
• 3D imaging

Innovation

• Plug-and-Play
• Compressive Sensing Light-Sheet Microscopy
• Axial-Coupled Model
• Woodbury Update
• Volumetric Reconstruction