pycudasirecon

This package provides a python wrapper and convenience functions for cudasirecon, which is a CUDA/C++ implementation of Mats Gustafsson & Lin Shao’s 3-beam SIM reconstruction software. [1]. It also supports lattice-light-sheet SIM (with deskewing).

Additionally, it includes a GPU implementation of the Hessian denoising algorithm described in Huang 2008 [2].

  • 3D (3-beam) SIM reconstruction

  • 2D (2-beam) SIM reconstruction, including TIRF-SIM

  • Lattice-SIM reconstruction

  • GPU-backed Hessian denoising

Install

Install (Linux and Windows) from conda forge:

conda install -c conda-forge pycudasirecon

see GPU requirements in Installation.

Quickstart

If you have a PSF and an image volume and you just want to get started, check out the pycudasirecon.reconstruct() function, which should be able to handle most basic applications.

from pycudasirecon import reconstruct

raw = tf.imread('path/to/raw_data.tif')
psf = tf.imread('path/to/sim_psf.tif')
makeotf_kwargs = {}  # kwargs for pycudasirecon.make_otf
recon_params = {}  # kwargs for pycudasirecon.ReconParams
result = reconstruct(
    raw,
    psf=psf,
    makeotf_kwargs=makeotf_kwargs,
    **recon_params
)

This library is in development … more details to follow.

References

[1]

Mats G.L. Gustafsson, Lin Shao, Peter M. Carlton, C. J. Rachel Wang, Inna N. Golubovskaya, W. Zacheus Cande, David A. Agard, and John W. Sedat. Three-dimensional resolution doubling in wide-field fluorescence microscopy by structured illumination. Biophysical Journal, 94(12):4957–4970, June 2008. URL: https://doi.org/10.1529/biophysj.107.120345, doi:10.1529/biophysj.107.120345.

[2]

Xiaoshuai Huang, Junchao Fan, Liuju Li, Haosen Liu, Runlong Wu, Yi Wu, Lisi Wei, Heng Mao, Amit Lal, Peng Xi, Liqiang Tang, Yunfeng Zhang, Yanmei Liu, Shan Tan, and Liangyi Chen. Fast, long-term, super-resolution imaging with hessian structured illumination microscopy. Nature Biotechnology, 36(5):451–459, April 2018. URL: https://doi.org/10.1038/nbt.4115, doi:10.1038/nbt.4115.