Single Cell Classifier

Single Cell Classifier (SCC) is an ImageJ/Fiji plugin that aims to classify cells based on features related to their nuclei and their neighbors. The detection is done with the StarDist plugin.

This example show the classification (middle) of an H&E image (left) and the fluorescence control image (right).

Release

The release jar files can be found here.

Documentation and examples

The plugin documentation is available here.

You can find Python scripts to train your own models or analyze your measurements here. You can find the data used in these scripts here.

Existing models

The following models are included in the plugin:

Segmentation

• Nuclei Breast 20X H&E PDX: Nuclei segmentation for Breast 20X H&E PDX models.

Models from StarDist

• Versatile (fluorescent nuclei): Nuclei segmentation for fluorescence images.
• Versatile (H&E nuclei): Nuclei segmentation for H&E images.
• DSB 2018 (from StarDist 2D paper): Nuclei segmentation for fluorescence images trained from the 2018 Data Science Bowl dataset.

Classification

• Human/Mouse Breast 20X H&E PDX: Human/Mouse classification for Breast 20X H&E PDX models.
• Human/Mouse Breast 20X DAPI PDX: Human/Mouse classification for Breast 20X DAPI PDX models.