Computer-aided diagnosis of lung nodule using gradient tree boosting and Bayesian optimization

Source code of the following paper: Computer-aided diagnosis of lung nodule using gradient tree boosting and Bayesian optimization (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0195875)

We developed our software using the following softwares.
We recommend Anaconda distribution and package manager (pip) for preparing for these softwares.

• Ubuntu Linux (versio 16.04)
• shell (i.e., bash)
• Python (version 2.7, https://www.python.org/)
• scikit-image (version 0.18.1, http://scikit-image.org/)
• sciki-learn (version 0.18.1, http://scikit-learn.org/)
• xgboost (version 0.6, http://xgboost.readthedocs.io/en/latest/)
• hyperopt (version 0.1, http://hyperopt.github.io/hyperopt/)

CT images of lung nodules and corresponding labels obtained from LUNGx Challenges and NSCLC Radiogenomics are stored as NPY files in `training_data`. These files are visually verified by board-certified radiologists.

Please run the following command.

NPY files are licensed under Creative Commons Attribution 3.0 Unported License.
Code of this software is licensed under GNU GENERAL PUBLIC LICENSE versioin 3 or lator.
If NPY file or this code is used, please refer to our paper.

Code of our CADx system and binary data of lung nodules (36.1MB)
S1 File includes Python script of our CADx system and binary data of lung nodules stored as NPY. (ZIP)