Determination of NET Content
This package contains support material (code & data) for the paper:
Automatic Determination of NET (Neutrophil Extracellular Traps) Coverage in Fluorescent Microscopy Images by Luis Pedro Coelho, Catarina Pato, Ana Friães, Ariane Neumann, Maren von Köckritz-Blickwede, Mário Ramirez, and João André Carriço. Currently under review.
Upon publication, all the code will be made available under the MIT license. Use in academic publications should cite the paper above.
Original data is available in the directory
data/. The naming structure is
- prefix "image"
- nr of the sample
- nr of the field inside the sample
- channel (protein, dna, rois, or rois2).
For example, the file
image_25_00_protein.png is from image 25, index 0,
and is the protein channel. The ROI files are the result of human labeling.
See the manuscript for details on data acquisition.
The source code is split into two directories
netsthis is the library code, which is useful to adapt to new projects.
reproducein this directory, you will find all the necessary code to
reproduce all figures in the paper (including supplemental material).
- This is the central file which runs the whole analysis
- This script performs the reported comparison between the two operators
- This script evaluates Bernsen thresholding for different sets of parameters
- This builds a side-by-side Figure showing differences between operators.
- This draws composite images for all inputs images
Adapting to your own data
There is a detailed tutorial on how to use the library.
For running on your own data:
Additionally, for reproducing our experiments:
requirements.txt in the
source-code directory lists all the
requirements. If you have permission to do so, running the following command
inside that directory should install all dependencies:
pip install -r requirements.txt
If this fails, try:
sudo pip install -r requirements.txt
Reproducing the paper
The results of the paper can be reproduced on a Unix-like machine by running
reproduce.sh script inside
source-code/reproduce after having
installed the the requirements as listed above.
To use multiple processors, edit this script and set the value of the