Cell_Dist_Mesh_Generator

Introduction

This repository hosts the Cell_Dist_Mesh_Generator_V5.ijm, a FIJI/ImageJ macro tailored for the automated generation of distance meshes between cells, aiding in quantitative visualization.

Prerequisites

Before using this macro, ensure the following plugins are activated through the FIJI Updater:

• clij
• clij2
• clijx-assistant
• clijx-assistant-extension
• 3D ImageJ Suite (dependency of clijx-assistant-extension)
• PTBIOP (LaRoMe)
• IJPB-Plugins (MorphoLibJ)

1. Load the Cell_Dist_Mesh_Generator_V5.ijm macro in FIJI and click ‘Run’.
2. Select the input image folders and specify a destination folder for the output results.
3. Use single-channel, whole-cell signal images in TIF format for input.

Note: It is recommended to adjust the macro parameters and Scale_calibration_ratio according to your specific experimental images.

The macro initiates an automated batch image processing and exporting procedure, encompassing the following steps:

1. Pre-processing: Applies a Difference of Gaussian filter to reduce noise.
2. Segmentation: Utilizes the Huang threshold method (implemented in CLIJ2) for segmentation.
3. Mask Refinement: Processes the binary masks through ‘fill holes’, ‘opening box’, and ‘watershed’ operations, followed by connected component labeling.
4. Label Extension: Extends the labels using a Voronoi-like method until contact is made between them.
5. Post-Processing: Removes labels in contact with image edges and applies quality control filters, including size, Geodesic Elongation ratio, and Touching Neighbor counting, to minimize edge artifacts from the Voronoi-like extension.
6. Distance Mesh Generation: Employs the drawDistanceMeshBetweenTouchingLabels function from the CLIJ2 libraries to create the Distance Mesh.
7. Scale Calibration: Multiplies the resulting Distance Mesh image by the Scale_calibration_ratio to convert units from pixels to micrometers (µm).
8. Distance Mesh Dilation: Dilates the distance mesh using Morphological Filters in MorpholibJ to enhance visualization.

The macro outputs a Distance Mesh image, representing the spatial relationships between cells, which is crucial for quantitative cellular analysis.

1. FIJI:

   • Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., … & Cardona, A. (2012). Fiji: an open-source platform for biological-image analysis. Nature Methods, 9(7), 676-682. doi:10.1038/nmeth.2019

2. Huang Threshold Method (ImageJ / CLIJ):

   • Huang, L.-K., & Wang, M.-J. J. (1995). Image thresholding by minimizing the measures of fuzziness. Pattern Recognition, 28(1), 41-51. doi:10.1016/0031-3203(94)e0043-k

3. CLIJ2:

   • Haase, R., Royer, L. A., Steinbach, P., Schmidt, D., Dibrov, A., Schmidt, U., … & Myers, E. W. (2020). CLIJ: GPU-accelerated image processing for everyone. Nature Methods, 17, 5-6. doi:10.1038/s41592-019-0650-1
   • Vorkel, D., & Haase, R. GPU-accelerating ImageJ Macro image processing workflows using CLIJ. arXiv preprint.
   • Haase, R., Jain, A., Rigaud, S., Vorkel, D., Rajasekhar, P., Suckert, T., … & Myers, E. W. Interactive design of GPU-accelerated Image Data Flow Graphs and cross-platform deployment using multi-lingual code generation. bioRxiv preprint.

4. MorphoLibJ:

   • Legland, D., Arganda-Carreras, I., & Andrey, P. (2016). MorphoLibJ: integrated library and plugins for mathematical morphology with ImageJ. Bioinformatics, 32(22), 3532-3534. doi:10.1093/bioinformatics/btw413

5. LaRoMe (LABEL Image to ROI function bundle with PTBIOP):

   • GitHub – BIOP/ijp-LaRoMe: Some useful functions to get Label from ROIs and vice versa, and more! https://github.com/BIOP/ijp-LaRoMe

For any issues, suggestions, or contributions, please open an issue or submit a pull request. Your feedback is invaluable in improving this tool.