Human Factors

The emergence and spread of infectious diseases, like COVID-19, are on the rise. Can you identify patterns between population density and COVID-19 cases and identify factors that could help predict hotspots of disease spread?

COEUS: COvid-19 Elevation on hUman Side

Summary

COEUS utilizes the space-based data of Earth observations before the pandemic to see the correlation between SDOH (e.g., population, building densities, socioeconomic conditions and exposure to pollutants) and the real-time distribution of COVID-19 cases. With the recognized patterns, COEUS can further identify the possible indicators that could help to predict the spread and severity of the disease.

How We Addressed This Challenge

"Space apps challenge" is a contest of high expectations. For this reason, our team decided to work accordingly. Having as the main purpose "to identify patterns between human activity and COVID-19 cases and factors that could help predict hotspots of disease spread" we chose to tackle the challenge by using data from space-based assets before the pandemic in order to see the correlation between physical determinants and the real-time COVID-19 data, as we have described in our project slides. We identified as the first factor the NIGHT LIGHTS. We got advantage of the data provided by NOAA. Through the satellite image of  light pollution we are pleased to spot the places with a high concentration of people and also socio-economic status differences across areas which emit less or more light. Furthermore, AIR QUALITY is the second factor we identified and is really important to us. Utilizing the data from NASA Nitrogen Dioxide, Carbon Monoxide and Ozone maps we succeeded in recognizing an important factor which leads to  the air pollution so in the same time to respiratory problems. In these ways we make clear how NASA, ESA, JAXA, CNES and CSA datasets and resources can be used to face the challenge. Our project is rendered as a pioneering effort which tries to incorporate for the first time new ideas such as the consideration of light emission/pollution.

https://covidcoeus.wordpress.com/


How We Developed This Project

     Health is our most precious possession. To avoid COVID-19 or other viruses spread, humanity should know the mechanisms that predict hotspots of disease spread and also to identify patterns between population density and COVID-19 cases. Our team chose to be on this side of the virus treatment. On the side of prediction and recognition of those patterns that lead to pandemics. Moreover, as soon as the pandemic started, we saw a huge difference in the spread and severity in different nations, especially the difference between Asian and Westerns countries. Such a huge difference motivated us to find the possible factors which resulted in the whole world in the present situation. Are the factors purely human/social or does nature also play a significant role? This specific question fascinated us and we desired to figure it out with the aid of open data.     

       Our main goal was to make full use of the data provided by NASA, ESA, JAXA, CNES, and CSA. This is the reason why we chose to mainly include our project scientific map data from NASA. Another very important tool that we took into consideration was machine learning. In a world that computational technologies are constantly developing it was almost impossible not to take advantage of the helpful machine learning and data visualization. The first step to our effort was to find the appropriate data in order to decide on our idea.The span of January 2020 was taken into consideration for data. Then, after the consideration of the valuable information, we started “building” our proposal to the challenge. We concentrated on different atmospheric maps in order to see if air pollution plays a significant role in the virus spread and effects. Lastly, we thought about a feature that has never been included as a hotspot prediction in people’s thinking: Light Pollution. Python,OpenCV was used for detecting the brighter areas in the image,Google earth engine was used for data processing and visualization.

       The truth is that we faced a lot of difficulties as a team in order to complete the project. One of them was the lack of direct cooperation and communication. Most of the members do not come from the same country or even continent, so it was quite difficult due to the different time zones to reach an agreement. Secondly, one big challenge for us was to find both innovative and new ideas that will provide a previously unknown method for identifying the patterns. On the contrary,even though we were from different regions,we enjoyed being in a team.We utilized the time difference as different shifts for different person to work.It was full of cultural diversity as well because we speak different languages.Even though we faced difficulties ,existed the spiritual pleasure and satisfaction that research and teamwork offered us. It is a great experience for all of us.

This is our website https://covidcoeus.wordpress.com/

Project Demo
  • Please see our slides as attached:

COEUS: COvid-19 Elevation on hUman Side

  • And the project code:

COEUS

Data & Resources

1. Night light data

NASA: VIIRS nighttime imagery in Worldview

VIIRS Stray Light Corrected Nighttime Day/Night Band Composites Version 1 by NOAA : https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG?authuser=1

2. Air quality data


Tags
#air quality, #night light, #age, #SDOH, #machine learning #space #NASA #getinspired #ESA #JAXA #CSA #CNES#NOAA#Python#humanfactors#innovation#challange
Global Judging
This project was submitted for consideration during the Space Apps Global Judging process.