Skip to content

Latest commit

 

History

History
88 lines (61 loc) · 2.64 KB

README.md

File metadata and controls

88 lines (61 loc) · 2.64 KB

Team 9 SCCM Datathon 2023 - COVID Trajectories Analysis

Code style: black

Table of Contents

Introduction

Here is our code to run experiments to analyse Social Determinants of Health associated with COVID-19 trajectory.

If you find this project interesting, we would appreciate your support by leaving a star ⭐ on this GitHub repository.

Author: Adrien Carrel, MSc, Meng; Tien "Amy" Bui; Yugang Jia; Lasse Hansen; Damien Archbold; Ivor S. Douglas, MD, FRCP (UK); Peter E. Morris, MD

Installation

You can clone or fork this repository to your local machine.

git clone https://github.com/SCCMdatathon2023/team_09

Dependencies

This project requires the following dependencies:

  • fastdtw>=0.3.4
  • hdbscan>=0.8.33
  • kaleido>=0.1.0.post1
  • matplotlib>=3.7.2
  • nltk>=3.8.1
  • numpy>=1.24.4
  • pandas>=2.0.3
  • plotly>=5.15.0
  • scikit_learn>=1.3.0
  • scipy>=1.11.1
  • tableone>=0.8.0
  • tqdm>=4.65.0
  • ~aleido>=0.2.1

Please make sure you have the required dependencies installed before using the code in this project.

You can install all of them by running the command:

pip install -r requirements.txt

Some other packages may be required.

Usage

To use our project or replicate the results, run our different python script or you can simply run the Python commands individually. For example, for the timeseries clustering project:

config = {
    # your parameters here
}
project = Team9(**config)
project.run_clustering()
project.analyze_clusters()

Citation

If you use piece of code from this project in your research or work, please consider citing it using the following BibTeX entry:

Carrel, A., Bui, T., Jia, Y., Hansen, L., Archbold, D., Douglas, I. S., & Morris, P. E. (2023). Analysis of SDOH and COVID-19 Trajectories (Version 1.0.0) [Computer software]. https://github.com/SCCMdatathon2023/team_09

Acknowledgement

SCCM Discovery Datathon 2023, the mentors and the organizers for helping us out during these two days.

License

This project is licensed under the MIT License. Feel free to use and modify the code as per the terms of the license.