This project has been archived!!

The deidentification features have been moved to the deidcm project The rest of the features have been included into the deep.piste project

kskit module

Functionality for cancer screening data pipeline including DICOM image importing and processing.

Initially conceived for french breast cancer screening program during the execution of deep.piste study

Documentation

kskit documentation can be found at: https://epiconcept-paris.github.io/kskit/

Installation

pip install kskit

Installation for contributors

1. Download source code

git clone https://github.com/Epiconcept-Paris/kskit.git
cd kskit

2. Create and activate a virtual environment

python3 -m venv env
. env/bin/activate

3. Install kskit

pip install -e .

Checking installation

Open a python interpreter and try to deidentify a dicom file:

from kskit.dicom.deid_mammogram import deidentify_image_png

deidentify_image_png(
    "/path/to/mammogram.dcm",
    "/path/to/processed/output-folder",
    "output-filename"
)

Tools for developers

Installation

pip install -e .[quality-tools]

Usage

Format your files with python3 -m autopep8 --in-place file/to/format

Lint your files with python3 -m pylint file/to/lint

Run Tests

Run all tests

pytest

Run a specific test file

pytest test/test_df2dicom.py

Run all except OCR tests

pytest --ignore=test/test_ocr_deidentification.py --ignore=test/test_df2dicom

Show full error message

pytest test/test_df2dicom.py --showlocals

Calculate Tests Coverage

1. Produce the .coverage file

coverage run --omit="*/test*" -m pytest

2. Visualize the coverage report in the terminal

coverage report -i

3. Produce an HTML report with test coverage

(The report will be available in htmlcov/index.html )

coverage html -i

Documentation

Run development server

mkdocs serve

Deploy documentation to GitHub Pages

mkdocs gh-deploy