The aim of this project is to classify tumour cells using "Giga images". To do this, we cut up the image to process (parallelizable) tiles. Then we use a deep learning algorithm to segment the cells and a simple algorithm for segmentation.
First, create a new environment and acivated it :
python3 -m venv env
source env/bin/activate (Unix/macOS)
.\env\Scripts\activate (windows)
Then, intalled the required librairies :
pip install -r requirements.txt
python main.py [-s|--slide_folder]
MANDATORY:
- [-s|--slide_folder] Path to the folder cointaing the slides. Could be avoid by changing the DEFAULT_INPUT_SLIDE_FOLDER in parameters.py.
The outputs are the following :
- saved plot in plot/Classification/"slide_name"/
We are trying to classify the tumoral cells, between 0,1,2,3. The plots show the patch, the classifcation and some cells of each class.