An Efficient Inference Frame for SMLM (Single-Molecule Localization Microscopy)

TL;DR

This paper introduces DilatedLoc, a lightweight neural network for SMLM that improves inference speed and image quality while maintaining a small model size, facilitating practical deep learning applications in microscopy.

Contribution

The paper presents a novel, efficient deployment framework and a lightweight neural network, DilatedLoc, optimized for fast and high-quality SMLM image reconstruction.

Findings

DilatedLoc reduces model size to under 100 MB.

Achieves 50% faster inference speed compared to existing models.

Provides superior GPU utilization with a new deployment architecture.

Abstract

Single-molecule localization microscopy (SMLM) surpasses the diffraction limit, achieving subcellular resolution. Traditional SMLM analysis methods often rely on point spread function (PSF) model fitting, limiting the application of complex PSF models. In recent years, deep learning approaches have significantly improved SMLM algorithms, yielding promising results. However, limitations in inference speed and model size have restricted the widespread adoption of deep learning in practical applications. To address these challenges, this paper proposes an efficient model deployment framework and introduces a lightweight neural network, DilatedLoc, aimed at enhancing both image reconstruction quality and inference speed. Compared to leading network models, DilatedLoc reduces network parameters to under 100 MB and achieves a 50% improvement in inference speed, with superior GPU utilization through a novel deployment architecture compatible with various network models.