The tool detects and counts the neurons and the neurons with satellite cells. The first channel is supposed to contain a staining of the neurons and the second channel a staining of all nuclei (neurons and satellite cells).

You can find the tool's source code here.

You must have the plugins 3D ImageJ Suite [1] and MorphoLibJ [2] installed. They are available via the FIJI-update-sites 3D ImageJ Suite and IJPB-plugins.

To install the tool save the file count_satellites_tool.ijm into the folder macros/toolsets of your FIJI installation.

Select the "count_satellites_tool" toolset from the >> button of the ImageJ launcher.

• the first button (the one with the image) opens this help page
• the c-button counts the satellites on the current image

Both channels are pre-processed. A blurred version of the image is subtracted from the original image and a Gaussian-Blur-filter with a small radius is applied to the result. The nuclei-channel is thresholded using the Triangle-auto-thresholding. A 2d watershed and a 3D-watershed are applied to the mask to separate touching nuclei. The result is eroded (3D) and the nuclei are detected with the help of the 3D Objects Counter. The neuron-image is thresholded using the Triangle-auto-thresholding method. Nuclei are labelled as belonging to neurons if the resulting mask is bigger than zero at the center of the nuclei and as possible satellites if the mask is zero. The minimal distances between the centers of the neurons and the possible satellites are calculated. Each possible satellite that is closer than a given distance to a nucleus is counted as an actual satellite cell.

Right-click the c-button to open the options-dialog.

nuclei thresholding method

The auto-thresholding method used to segment the nuclei in the DAPI-channel.

nr. of erosions

The number of erosions applied to the mask of the nuclei.

min. size

Objects with a smaller number of voxels are excluded from the result.

max. size

Objects with a larger number of voxels are excluded from the result.

neuron thresholding method

The auto-thresholding method used to segment the neurons in the NeuN-channel.

marker style neuron

The style of the rois indicating the detected neurons.

marker style satellite

The style of the rois indicating the detected satellite cells.

Rois indicate the detected neurons and possible satellite cells (nuclei not belonging to neurons). A channel with ellipsoids indicates the pairs of detected neurons and satellite cells. Each pair has a unique color-index.

Two tables are created. One displays the number of neurons and satellite cells per image. The other lists the detected pairs of neurons and satellite cells and reports the ids (corresponding to the roi labels) of the cells, their coordinates, the distance between neuron and satellite cell and the volume of the satellite cell.

• Use the 3d-viewer on the result image.
• Use the channels-tool (Image>Color>Channels Tool) to switch on and of the channels in the result image.
• Use Image>Overlay>Labels to switch on and of the labels of the rois. The labels correspond to the ids in the result-table.
• If the input image has a selection, the analysis is run on the selected area. For big images it could be useful to select the region containing the cells.

[1] Ollion, J., Cochennec, J., Loll, F., Escudé, C., and Boudier, T. (2013). TANGO: a generic tool for high-throughput 3D image analysis for studying nuclear organization. Bioinformatics 29, 1840–1841.

[2] Legland, D.; Arganda-Carreras, I. & Andrey, P. (2016), "MorphoLibJ: integrated library and plugins for mathematical morphology with ImageJ", Bioinformatics (Oxford Univ Press) 32(22): 3532-3534, PMID 27412086, doi:10.1093/bioinformatics/btw413 (on Google Scholar).