Greenojo - Space Apps Challenge

Greenojo| Where There’s a Link, There’s a Way

Where There’s a Link, There’s a Way

Since the COVID-19 pandemic began, there has been a proliferation of websites and portals developed to share resources about the topic. Your challenge is to find innovative ways to present and analyze integrated, real-time information about the environmental factors affecting the spread of COVID-19.

Covid-19 Risk Spread correlation with Climate

Summary

a) We found that an increase of x% in each of these parameters (from Air Quality, Weather and Radiation) increases the COVID-19 death rate by y%, while y > x in most of the cases with 98+% confidence intervalb) We also created 3 training models by taking 2 months of data (March and April) and tried to predict for cases at each county level by taking Environmental datasets for the month of May as the test data. Then we did the variance calculation on the actual/test data, which gave us a variance less than 5%c) The study results indicate the positive correlation of “Confirmed, Deaths, Fatality Rate” with the hybrid environmental data -“Air Quality + Basic Weather Parameters + Radiation")

How We Addressed This Challenge

Our study establishes potential link between human health (covid-19/lung infection risk spread) and the environment.

We could build evidence supporting causal (versus correlational) relationships between environmental observations and the pandemic spread.

Based on integrated data, we can do reliable predictions about the logistics of healthcare resources allocation.

How We Developed This Project

To contribute in terms of supporting global policymakers. We took this as a Machine Learning project and tried to demonstrate by building training models for risk prediction. We used Matlab and Python for running Data Analytics, Tableau for visualization, NVIDIA GPU processors for training models, Google Colab for running Jupyter Notebooks. We have used various data sources such as AirNow (for Air Quality), meteomatics.com (for basic weather data) and epa.com (for radiation data). Our team was able to combine multiple forecasting models and performed optimization on the training dataset.

Project Demo

https://drive.google.com/open?id=1CzsK433y9MrqfTsJ15ZXZnjwyGf5ycl9

Project Code

N/A

Data & Resources

We have used various data sources such as AirNow (for Air Quality), meteomatics.com (for basic weather data) and epa.com (for radiation data).

Global Judging

This project was submitted for consideration during the Space Apps Global Judging process.