Automatic Cell Counter

What is this?

This repository provides an automatic cell counter algorithm to count cell numbers in 2D microscopy images from Boyden Chamber assay and offers a user interface for further manual correction.

Please star this repository if you like it :)

Installation:

This algorithm has been implemented using Python 3.9. To install the necessary packages for this framework run:

pip install -r requirements.txt

If you are using conda first install pip by: conda install pip

Usage:

To run the cell counter, the user can easily input the following code in the terminal.

Require arguments:

• -image: The path of the input image or the path of the directory.

python viewer.py --image example_data/8.tif

or

python viewer.py --image example_data/

Here is a tutorial to set up your Automatic Cell Counter for Windows: Tutorial

Example output:

The user interface will be built based on every input image for manual correction. Inside each image, red dots are used to denote the centers of the detected cells. We also used green dots and red bounding boxes to guide users to pay more attention to those easy-to-make-mistake areas: green dots (detected cells are quite small, perhaps one cell is divided to several parts) and red bounding box (high density areas, perhaps two cells are merged). The user may add dots or delete dots manually based on the automatically detected results.

After manual correction, press the 'd' (denotes Done) key on the keyboard to print the updated cell numbers on the terminal and close the current window. When all images are corrected, an excel file named result will be generated which contains both automatically detected cell numbers and corrected cell numbers.

User interface for manual correction:

Example output on the terminal (left) and in the excel (right):

Data:

The target images for this reporsitory are taken from Boyden Chamber assay under the microscope – Brightfield, 20x. The assay is based on a chamber of two medium-filled compartments separated by a microporous membrane. The cells are placed on the top of the chamber and are allowed to migrate/invade through the pores of the membrane into the lower compartment. After a certain time, the cells are fixed and stained with 0.05% Crystal Violet – and the number of cells into the lower compartment are counted to determine the migratory or invading capacity of the cells.

The parameters of this algorithm are determined for image with size of 1550 * 2088. If your images are much smaller or larger than this size, please adjust the parameters.

Example images:

Methods:

How to cite:

@software{ruolin_shen_2022_6390980,
  author       = {Ruolin Shen and
                  Christina Bukas and
                  Marie Piraud},
  title        = {{HelmholtzAI-Consultants-Munich/Automatic-Cell- 
                   Counter: First version of the software}},
  month        = mar,
  year         = 2022,
  publisher    = {Zenodo},
  version      = {v1.0.0},
  doi          = {10.5281/zenodo.6390980},
  url          = {https://doi.org/10.5281/zenodo.6390980}
}