EpidemicSimulationSandbox

Passion Project for Advanced Computer Science at the Mass Academy of Math and Science made by Atharv Joshi and Arnav Prabhudesai

How to run the frontend

In the terminal, run:

cd v-0.2-earthdata/my-react-app

npm i

npm run dev

Copy and paste the url that it gives you (ex: localhost: 5173)

Enjoy!

How to run the transfer system

Open v-0.2-earthdata/backend/sender.html

Input all the data it asks for (leave the Server URL field as the default value)

Open v-0.2-earthdata/backend/receiver.html

Wait for the server to connect. If the server has disabled due to inactivity, this could take up to a minute. Accept the download of the first frame.

How to run the simulator

Run v-0.2-earthdata/EarthdataSimulator.m in MATLAB. The frames will be outputted at the location of v-0.2-earthdata/backend/sim_frame.bin

The output file will be in the format of a 3MB binary file which represents a 720x1440 image with one byte for red, one byte for green, and one byte for blue.

Project Overview

Our model accurately simulates virus spread for a certain set of parameters. It has a high amount of customizability as it allows user modification of parameters for virus spread. The model visually represents the results by a world map UI that shows the impact of the virus where a larger red circle means more people are infected, aand a more red circle means more people are infected

The system follows a SEIRDS model (Susceptible -> Exposed -> Infected -> Recovered/Dead -> Susceptable)

So far, the system has these parameters:

• Infectivity
• Deadliness
• Heal Rate
• Immunity Loss Rate
• Severity
• Spreadability

In addition, the system also uses real population data, gathered by NASA in 2020.

Documentation can be found at: http://sedac.ciesin.columbia.edu/data/collection/gpw-v4/documentation

More data can be found at: http://sedac.ciesin.columbia.edu/data/collection/gpw-v4/sets/browse

Next Steps

After building a successful MVP, here are our next steps to make the project more realistic, and to turn it into a deployable solution.

• Implement the ability to travel in the simulation

• Clairy and configure the server connecting the backend --> frontend (doesn't work exactly how we want it to right now... The system downloads the first frame as a MVP of the transfer system. It does not yet provide a steady 10fps stream of binary simulation data).

• Implement the ability to drop the virus onto the simulation... right now, the virus just spawns in Lebanon!!

• Implement connectivity from the frontend --> backend, so that changing sliders actually changes the sim.

• Create better/more accurate preventative measures in place of our blanket "vaccinate" button in MATLAB which just turns 75% of the population into Recovered people

• Add realistic "mutation rate" that can be customized by the user

• Add new and more interesting virus shapes and colors and different virus types

• Add slider for the speed of the simulation

Then we'll have a finished product!

Let us know if you have any concerns or questions!

Best, Atharv and Arnav