n2c2 2018 — Track 1: cohort selection for clinical trials

This repository contains source code from our participation in the n2c2 2018 shared-task (Track 1).

Requirements

Ubuntu, Python 3.6.4. Install the required packages:

$ pip install -r requirements.txt

Usage

1. Place the n2c2 dataset into the data/n2c2/ directory.

2. Input parameters can be modified in the src/system.py script.

3. In the src/ directory run:

   $ python3 system.py
   

Reference

If you use this code in your work, please cite our publication:

@inproceedings{antunes2019a,
  address   = {{Prague, Czech Republic}},
  author    = {Antunes, Rui and Silva, Jo{\~a}o Figueira and Pereira, Arnaldo and Matos, S{\'e}rgio},
  booktitle = {12th {{International Joint Conference}} on {{Biomedical Engineering Systems}} and {{Technologies}}},
  month     = feb,
  pages     = {59--67},
  publisher = {{SciTePress}},
  title     = {Rule-based and machine learning hybrid system for patient cohort selection},
  url       = {https://doi.org/10.5220/0007349300590067},
  year      = {2019},
}