Analytical reference framework to analyze non‑COVID‑19 events

Impact of Covid-19 in Colombia

Development and evaluation of mathematical and epidemiological models that support decision-making in response to the Covid-19 emergency in Colombia. The project was approached from the perspective of data science, using data analytics and machine learning techniques.

Papers

This work has been published as an article in the journal Population Health Metrics. A draft of the paper can be found here and the online version here.

Framework

The framework proposed and implemented in this work, named “Analytics for Non-COVID-19 Events” (ANE, from now on), adapts the ASUM-DM (Analytics Solutions Unified Method for Data Mining) methodology to include and handle specific characteristics of health events and their underlying data. The proposed framework can be seen in the following image.

Implementation

The proposed analytical framework was mainly implemented in 2 software components:

1. Descriptive Engine
2. Predictive Engine

Events / Diseases

The selected events to which we applied the proposed framework are:

1. Tuberculosis (TB)
2. Suicide Attempt (SA)
3. Infant Mortality (IM)
4. Diabetes Mellitus (DM)
5. Acute Diarrheal Disease (EDA)
6. Excess Mortality (EM)

Click here to view the dataset files for the descriptive solution.

Click here to view the dataset files for the predictive solution.

Data

Primary sources:

• SIVIGILA
• SISPRO
• DANE
• Colombia Open Data

The data (both source and predictions) are up to date:

• Tuberculosis: 12/27/2020
• Suicide Attempt: 12/27/2020
• Infant Mortality: 4/12/2020
• Diabetes Mellitus: 4/12/2020
• Acute Diarrheal Disease: 4/12/2020
• Excess Mortality: 4/12/2020

Dependencies

The project was carried out with the latest version of Anaconda on Windows.

The program can also be run on Linux with Python 3.7.x (or higher) by previously installing the following libraries:

  sudo apt install
  sudo add-apt-repository universe
  sudo apt-get update
  sudo apt install python3-pandas
  sudo apt install python3-sklearn
  sudo apt install python3-statsmodels

Author

• Created by Andrés Segura Tinoco
• Created on Jul 15, 2020
• Last update on Sep 06, 2021

Acknowledgements

We would like to make a special acknowledgement to the National Health Institute of Colombia.