README.md

SACDj

Fater super-resolution fluctuation imaging: SACD reconstruction with FIJI/ImageJ.

v1.1.3

This repository is for Simplified SACD (w/o Sparse deconvolution) and will be in continued development. It is a part of publication. For details, please refer to: Weisong Zhao et al. Enhanced detection of fluorescence fluctuation for high-throughput super-resolution imaging, Nature Photonics (2023). Please cite SACD in your publications, if it helps your research.

Portal to the plugin.

The related MATLAB version can be found at HERE

You can also find some fancy results and comparisons on my website.

If you are interested in our work, I wrote a #behind_the_paper post for further reading.

SACD reconstruction

SACD demo

These two demos can be found at the release v1.1.3.

Instruction

• Tips: Regarding the SACD SR frame visualization, it can be scaled with a gamma correction according to the bSOFI setting.

In FIJI/ImageJ: 
	Process->Math->gamma(0.5)
	Process->Filters->Gaussian Blur(1)
	Process->Enhance Contrast(0%; normalize)
Macro:
run("Gamma...", "value=0.50");
run("Gaussian Blur...", "sigma=1");
run("Enhance Contrast...", "saturated=0 normalize");

• Tips: If data contains strong background, a pre background subtraction will help.

In FIJI/ImageJ: 
	Process->Subtract background

Declaration

This repository contains the java source code (Maven) for SACD imagej plugin. This plugin is for the Simplified SACD (w/o Sparse deconvolution), and is also accompanied with conventional SOFI reconstruction ; RL deconvolution; and PSF calculation features. The development of this imagej plugin is work in progress, so expect rough edges.

If you want to reproduce the results of SACD publication, the SACDm (Matlab version) is recommended. Due to the distance between the Fourier interpolation, deconvolution of SACDj, and SACDm, there may exist a gap between the results of SACDm and SACDj. For me, the implementations of SACDm are more flexible and accurate.

Version

• v1.1.3 Fixed for memory recycle & 20 times execution acceleration
• v1.1.0 Useful tools separated: RL deconvolution, PSF calculation, SOFI reconstruction & Fourier interpolation
• v1.0.0 Simplified SACD
• v0.6.0 Accelerated RL-TV deconvolution
• v0.5.0 Accelerated RL deconvolution
• v0.4.0 Born-Wolf PSF
• v0.3.0 SACD core
• v0.2.0 Fourier interpolation
• v0.1.0 Cumulant reconstruction core

Related links:

• MATLAB version of SACD: SACDm
• Some fancy results and comparisons: my website
• Preprint: Weisong Zhao et al. Enhancing detectable fluorescence fluctuation for high-throughput and four-dimensional live-cell super-resolution imaging, bioRxiv (2022).
• Reference: Weisong Zhao et al. Enhanced detection of fluorescence fluctuation for high-throughput super-resolution imaging, Nature Photonics (2023)

Plans

• Improve the perfomance of Fourier interpolation;
• Remove redundant code and reconsitution ugly code. (in progress)
• Accelarated RL deconvolution.
• Accelarated RL-TV deconvolution.
• Add sparse deconvolution.
• Automation

Open source SACDj

This software and corresponding methods can only be used for non-commercial use, and they are under Open Data Commons Open Database License v1.0.