COVID-19 Spread Predictor

Case data for the US and a fit logistic model. Epidemic stages are shown in different colors representing the type of growth.

Inspiration

Inspired to put my engineering skills to good use, I created a program that models the spread of COVID-19 in any country around the world. I wanted to make sure that the program was able to provide simple, robust, and accurate predictions regarding case data of COVID-19 in any country.

What it does

This program receives case data over the web for all countries from the Humanitarian Data Exchange. For a specified country, the data is fitted to a logistic model and then forecasting is displayed. The model can be used to predict the number of cases and deaths in a particular country during a short-time period. It also can be used to predict the peak of the epidemic in a certain country. The model does not describe all situations and therefore a statistical analysis should be performed after using.

How I built it

Inspired by the work of Milan Batista, I was able to build a robust and dynamic code that fits a logistic model for nearly any country based on data from the HDX. This is done using MATLAB and the Statistics and Machine Learning package.

Challenges I ran into

It was really hard obtaining the data over the web and processing it to be compatible with the mathematical structure.

Accomplishments that I'm proud of

I'm really proud that I was able to make this dynamic code possible and to share it around the world. I've received great feedback from many researchers and peers who find my code to be a great, accurate, and easy to use predictor of COVID-19 spread.

What I learned

I learned that it is possible to complete a seemingly difficult project with the help of others. I collaborated with people around the globe after publishing this project and it was really inspiring to learn how much others care about helping humanity.