Particle detection
Isolated object detection



This is an implementation of Mask R-CNN on Python 3, Keras, and TensorFlow. The model generates bounding boxes and segmentation masks for each instance of an object in the image. It's based on Feature Pyramid Network (FPN) and a ResNet101 backbone.

Nuclei Segmentation (Mask-RCNN)


NEUBIAS-WG5 workflow for nuclei segmentation using Mask-RCNN. The workflow uses Matterport Mask-RCNN. Keras implementation. The model was trained with Kaggle 2018 Data Science Bowl images.

has topic
need a thumbnail

Nuclei Segmentation (Python)


This workflow processes images of cells with discernible nuclei and outputs a binary mask containing where nuclei are detected.

need a thumbnail



The software FishInspector provides automatic feature detections in images of zebrafish embryos (body size, eye size, pigmentation). It is Matlab-based and provided as a Windows executable (no matlab installation needed).

The recent version requires images of a lateral position. It is important that the position is precise since deviation may confound with feature annotations. Images from any source can be used. However, depending on the image properties parameters may have to be adjusted. Furthermore, images obtained with normal microscope and not using an automated position system with embryos in glass capillaries require conversion using a KNIME workflow (the workflow is available as well). As a result of the analysis the software provides JSON files that contain the coordinates of the features. Coordinates are provided for eye, fish contour, notochord , otoliths, yolk sac, pericard and swimbladder. Furthermore, pigment cells in the notochord area are detected. Additional features can be manually annotated. It is the aim of the software to provide the coordinates, which may then be analysed subsequently to identify and quantify changes in the morphology of zebrafish embryos.

FishInspector Logo

Quantification of outer ring diameters of centriole or PCM proteins of cycling HeLa cells in interphase


This workflow can be ran with data from 3D-SIM showing the centrosomes in order to compare the distribution of diameters of rings (or toroids) of different proteins from the centrioles or the peri centriolar material. It aims to reproduce the results of the Nature Cell Biology Paper Subdiffraction imaging of centrosomes reveals higher-order organizational features of pericentriolar material  from the same data set but with a different analysis method.

It is slightly different from the methods described in the paper itself, where the method was to work on a maximum intensity projection of a 3D-SIM stack, and then to fit circle to the centrioles to estimate the diameters of the toroids.

In this workflow, the images are read from the IDR , then process by thresholding (Maximum entropy auto thresholding with Image J), and processed by Analyze Particles  with different measurement sets, including the bouding box. Then the analysis of diameters and the statistical test are performed using R. All the code and data sets are available, and in the case of this paper have shown a layered organisation of the proteins.

Combined view from Figure 1 Lawo et al.

ATLAS Vesicle segmentation method


Part of ATLAS software

Comment / Instructions: 

You can upload your image at the Mobyle@SERPICO portal and download the result. The workflow is only available online, i.e. no download possible.



This plugin tags all pixel/voxels in a skeleton image and then counts all its junctions, triple and quadruple points and branches, and measures their average and maximum length.


he tags are shown in a new window displaying every tag in a different color. You can find it under [Plugins>Skeleton>Analyze Skeleton (2D/3D)]. See Skeletonize3D for an example of how to produce skeleton images.

The voxels are classified into three different categories depending on their 26 neighbors: - End-point voxels: if they have less than 2 neighbors. - Junction voxels: if they have more than 2 neighbors. - Slab voxels: if they have exactly 2 neighbors.

End-point voxels are displayed in blue, slab voxels in orange and junction voxels in purple.

Notice here that, following this notation, the number of junction voxels can be different from the number of actual junctions since some junction voxels can be neighbors of each other.


Output data type: table result, image of the skeleton

Image removed.


Clustered cell nuclei


One of the principal challenges in counting or segmenting cells or cell nuclei is dealing with clustered objects. To help assess algorithms' performance in this regard, synthetic 3D image sets of HL60 cell line are provided consisting of four subsets with increasing degree of clustering. Each subset is also provided in two diferent levels of quality: high SNR and low SNR.Ground truth is available as well. The datasets are part of [Masaryk University Cell Image Collection (MUCIC)]( as well [Broad Bioimage Benchmark Collection (BBBC)]( - entry BBBC024.

has function
need a thumbnail

spot detection and codistribution analysis


WASH, Exo84, and cortactin spot detection and codistribution analysis To detect endosomes, an automatic Otsu threshold is applied to the Gaussian-filtered MT1-MMP–positive endosome image (= 1.5 pixels for the sample image). Statistics about each endosome are then saved, for example random positioning of spots can be compared to actual positioning. For each endosome, WASH and Exo84 (or WASH and cortactin) spots are searched for in a neighboring of x pixels in their respective channel. Their number and position are saved per endosome (**see the macro in Text file S2 downloadable from here**).

From the position of WASH and Exo84 (or WASH and cortactin) spots around each endosomes, each WASH spot is paired with its closest Exo84 (or cortactin) spot neighbor, optimized over all spots around this endosome.

This allowed measuring of the distribution of distance between WASH-Exo84 (or WASH-cortactin) spots (**for the co-distribution analysis, see matlab scripts in Zip file S3 downloadable).

endosomes and spot neighbors