This is an open-source python package supporting Radiomics medical image feature extraction and classification.
We aim to add a wide variety of features and classifiers to address a wide variety classification problems. Through a modular setup, these can easily be interchanged and compared.
For more information, see the sphinx generated documentation available here (WIP).
Alternatively, you can generate the documentation by checking out the master branch and running from the root directory:
python setup.py build_sphinx
The documentation can then be viewed in a browser by opening PACKAGE_ROOT\build\sphinx\html\index.html
.
PREDICT has currently only been tested on Unix with Python 2.7. The package can be installed through the setup file:
python setup.py install
Make sure you first install the required packages:
pip install -r requirements.txts
When running the FASTR package under version 1.3.0, you need to manually add the PREDICT fastr_tools path to the FASTR tools path. Go the your FASTR config file (default: ~/.fastr/config.py) and add the fastr_tools path analogue to the description in the PREDICT/fastrconfig/PREDICT_config.py file:
packagedir = site.getsitepackages()[0]
tools_path = [os.path.join(packagedir, 'PREDICT', 'fastr_tools')] + tools_path
When using FASTR >1.3.0, the PREDICT config file will be automatically created for you in the default: ~/.fastr/config.d folder.
Note that the Python site package does not work properly in virtual environments. You must then manually locate the packagedir.
We mainly rely on the following packages:
- SimpleITK (Image loading and preprocessing)
- numpy (Feature computation)
- sklearn, scipy (Classification)
- FASTR (Fast and parallel workflow execution)
- pandas (Storage)
See also the requirements file.
- We are working on improving the documentation.
- Examples and unit tests will be added.
This package is covered by the open source APACHE 2.0 License.
We are happy to help you with any questions: please send us a message or create an issue on Github.