diff --git a/README.md b/README.md index 9cad74eb1c504f4773fdf9ac309698097f2b229b..2b8fd55884bcf83999797293f16af8b0f338bc35 100644 --- a/README.md +++ b/README.md @@ -1,31 +1,94 @@ -# Zellige + +<img src="doc/zellige_pipeline.jpg" width="150" height="150" alt="drawing"/> +<img src= "doc/zellige_pipeline.jpg"> -## Workflow +# Zellige : 3D fluorescence microscopic images surface extraction tool. -The program takes a 3D stack as input and generates the projection of all surfaces contained in the 3D stack as 2D -images. That process can be divided into 3 stages : +Zellige is a software tool allowing the automated extraction of a non-prescribed number of various surfaces from a 3D +image acquired by fluorescence microscopy. The program is available through a FIJI plugin. -- first the selection of all the pixels belonging to any selectedPixels -- second the regrouping of those selected pixels to produce a height map for each selectedPixels -- third the use of each height map to extract a projection according to a given projection's method (only MIP for now). +**Table of content.** - +* [Citation.](#citation) +* [Motivation](#motivation) +* [Installation.](#installation) +* [Description.](#description) +* [Example datasets.](#example-datasets) +* [Usage.](#usage) -## Build ++ [Other projection tools.](#other-projection-tools) -```` +## Citation. - mvn clean install +If you use this work for your research, please cite the paper it is described in: + +> __Extracting multiple surfaces from 3D microscopy images in complex biological tissues with the Zellige software tool.__ +> +> Céline Trébeau, Jacques Boutet de Monvel, Gizem Altay, Jean-Yves Tinevez, Raphaël Etournay +> +> bioRxiv 2022.04.05.485876; doi: https://doi.org/10.1101/2022.04.05.485876 + +## Motivation + +The quantitative study of epithelium development by fluorescence microscopy requires either the dissection of the +epithelium to discard all potential contaminating structure (in terms of signal) surrounding it or the acquisition of +all the epithelium adjacent structures allowing to have a more accurate state of the epithelium in its biological and +physical environment. The latter approach requires to segment each structure separately. Most existing methods allow for +the extraction of a single surface witch has to be smooth, and presenting a sufficiently high signal intensity and +contrast and usually fail otherwise. These methods so far do not address the need to extract several surfaces of +interest within the same volume. We developed Zellige, a tool allowing the automated extraction of a non-prescribed +number of surfaces of varying inclination, contrast, and texture from a 3D image. The tool requires the adjustment of a +small set of control parameters, for which we provide an intuitive interface implemented as a Fiji plugin. + +## Installation. + +As a FIJI plugin (https://fiji.sc/) ,Zellige can be installed directly within +the [Fiji updater](https://imagej.net/ImageJ_Updater). + +In the `Manage update sites` window, check the `Zellige` plugin. Click the `Close` button, then the `Apply changes` +button. After the plugin is downloaded, restart Fiji. The plugin can then be launched from the _Plugins > Zellige_ menu +item. + +## Description. + +Zellige performs projection of multiple surfaces of interest on a 2D plane from a 3D image. For now, it can only be used +on 1-channel 3D images presenting one or several surfaces. The program takes a 3D stack as input and generates the +projection of all surfaces contained in the 3D stack as 2D images. That process can be divided into 3 stages : + + + +- first the selection of all the pixels belonging to any surface +- second the regrouping of those selected pixels to produce a **height map** for each surface +- third the use of each height map to extract a projection. + +The height-map is then used to extract a projection from the 3D image. A fixed offset can be specified separately for +each channel, and is used to collect intensity in planes above or below the reference surface. Several planes, specified +by a ∆z parameter, can be accumulated to generate a better projection, averaging the pixel values or taking the maximum +value of these planes. + +## Example datasets + +https://zenodo.org/record/6376584 +https://zenodo.org/record/6376594 +https://zenodo.org/record/6376582 +https://zenodo.org/record/6376566 +https://zenodo.org/record/6376542 - ```` - ## Usage -The plugin is currently in development -```` - -```` - -## Pipelines - -https://gitlab.pasteur.fr/ctrebeau/zellige-core/-/pipelines \ No newline at end of file +After installation, the plugin is located in the `Plugins >` menu of Fiji. The Zellige window is divided in 3 parts +corresponding to each step of the program. Its UI is a single window divided in 3 for each stage (selection, +construction and projection): + +### Other projection tools. + +There are several other tools to generate this kind of projections: + +- Local Z projector, an ImageJ plugin : +- Stack Focuser, an ImageJ plugin: https://imagej.nih.gov/ij/plugins/stack-focuser.html +- SurfCut, an Image macro: https://bmcbiol.biomedcentral.com/articles/10.1186/s12915-019-0657-1 +- PreMosa, a standalone software: https://academic.oup.com/bioinformatics/article/33/16/2563/3104469 +- Extended Depth of Field, an ImageJ plugin: http://bigwww.epfl.ch/demo/edf/ +- Min. Cost Z Surface, an ImageJ plugin: https://imagej.net/Minimum_Cost_Z_surface_Projection +- FastSME and SME, MATLAB software for the extraction of smooth-manifold + structures: https://openaccess.thecvf.com/content_cvpr_2018_workshops/w44/html/Basu_FastSME_Faster_and_CVPR_2018_paper.html