Project AD answers the question,“Is it safe to go to this area?”. By using our proposed website, citizens will be guided on what cities will be considered “safe” on a given day, especially for employees who commute daily into and out of Metro Manila cities. More importantly, by predicting COVID-19 incidence, local government officials can then make evidence-based decisions before restricting or allowing access to a city. Furthermore, the national government and policymakers are then able to predict which cities to deliver more healthcare resources.
We developed a website that predicts how much a chosen city on a day will pose a risk of COVID-19 exposure on the individual. Because there is a growing body of research showing that COVID-19 transmission is affected by the air temperature and humidity, and that air pollution worsens the severity of disease. Together with population data and the incidence of COVID-19 on a certain time period, we can then compute for the predicted incidence of COVID-19 of a city for a given day, this number we shall call “Hotspot percentage”.
Our project combines satellite data and local epidemiologic data regarding COVID-19 in order to provide information that can then be used by private and public entities which will serve as a guide in terms of transportation and logistics. Because of these websites, we concluded that there is a correlation between the virus and the environmental factors. Our reference page in our website gives the studies where we got our project idea.
If not working(projectareadenial.azurewebsites.net)
Inspiration:
There are two factors that inspired us to do this project, the first one is that due to lack of testing facilities, we cannot see where would COVID-19 spread next. “Silent” carriers or the ones that are not tested but has the virus can just show up anywhere without him knowing that he is contagious. Our country, the Philippines, has high population density in urban areas, particularly Metro Manila, where the number of COVID-19 cases is still rising. In order to limit the spread of this disease, widespread quarantine was imposed. Therefore, economic activity suffered greatly.
The second one is that our team is a former environmental research group that specializes air quality in a university at Manila, Philippines. Analyzing these environmental factors is one of our strengths, that is why we are confident that we can use this advantage to help people around the globe.
Approach:
We asked ourselves, is there a way to predict COVID-19 movement without actually testing the virus itself? We looked outside the box and hence, the birth of our project.
Space agency data:
We used satellite data to grab out some pictures or map out data if some environmental factors are present in our sample time and location. We also got raw data from various government agencies like NOAA and DOH.
NO2 data was from NASA's Goddard Space Flight Center.
Aerosol Thickness Data can calculate PM data. Can be retrieved from:
JAXA's MODIS / AQUA and TERRA Satellites.
1.) Choose what city the user wants to examine.
2.) Input the independent variables [Temperature, Relative Humidity, NO2, and PM10 values.]
3.) Click the "Submit" or calculate button.
Those input values will be used in a Random Forest Regression model made by our team. A "prediction" page will come out after submitting and will tell the user how many cases will happen given those variables.
4.) The higher the cases, the more contagious the virus is in that particular area.
We used Random Forest Regression since it can be used for classifying and predicting. It also good in handling "dirty" data since we used real-world data. Also, it has a good RMSE points compared to other models.
The tools used in this project are as follows:
Python programming:
- Data Analysis and Prediction [Pandas, Seaborn, Numpy, Matplotlib, Sklearn, Joblib] using Random Forrest Regression model.
- Flask, HTML and Bootstrap for our web design
Docker for images, Azure for Cloud Services, Porkbun for Domain Management
Problems Encountered:
Our expertise is around research and data analysis, but due to some time and monetary constraints our project has a weak point in UX, web design, web-scrapping real time data and the architecture of the website. As you will see in our website, a user will notice that we have to input the environmental factorssince we don't have the ability and the resources to web scrape real time data and plug it at the same time in our model. Also we would like to map out the calculations on a satellite map but we need more time and knowledge.
Achievements:
Only in this community quarantine we have encountered the term "Flask" and "Docker". Our biggest achievement would be that our website is running on an Azure cloud service with a custom domain name. We have some knowledge in data analysis in Python, QGIS, and HTML but actually setting up the website was a big success for us.
Trivia:
Why Project AD (Area Denial) ?: Area denial is a military tactic that uses strategy used to prevent an enemy from occupying an area of land. The tactic restricts, slows down, or endanger the opponent. (In this case, the virus)
Why Phi-6 (Feesiks)?: It's just we are physics majors and Phi is a Greek letter used in our Quantum mechanics subject.
References that are listed in the reference page of our website:
Raw data used for our model:
https://www.ncdc.noaa.gov/cdo-web/datatools/findstation
https://aqicn.org/data-platform/covid19/
https://www.doh.gov.ph/2019-nCoV
Satellite images to confirm correlation:
We used NASA Satellite for mapping NO2 images.
https://so2.gsfc.nasa.gov/no2/pix/htmls/Manila_data.html
We used JAXA Globe Portal System to have Aerosol Data
Satellites: MODIS / AQUA and Terra for Aerosol Thickness
Literatures:
Virus Literature:
Yuki, K., Fujiogi, M., Koutsogiannaki, S., COVID-19 pathophysiology: a review. Clinical Immunology, 215 (2020). https://doi.org/10.1016/j.clim.2020.108427
Ren, H., Zhao, L., Zhang, A., Song, L., Liao, Y., Lu, W., Cui, C., Early forecasting of the potential risk zones of COVID-19 in China's megacities. Science of The Total Environment, 729 (2020). https://doi.org/10.1016/j.scitotenv.2020.138995
Weather vs Virus Literature:
Gupta, S., Raghuwanshi, G.S., Chanda, A., Effect of weather on COVID-19 spread in the US: A prediction model for India in 2020. Science of The Total Environment, 728 (2020). https://doi.org/10.1016/j.scitotenv.2020.138860
Tosepu, R., Gunawan, J., Effendy, D.S., Ahmad, L., Lestari, H., Bahar, H., Asfian, P., Correlation between weather and Covid-19 pandemic in Jakarta, Indonesia. Science of The Total Environment, 725 (2020). https://doi.org/10.1016/j.scitotenv.2020.138436
Sobral, M., Duarte, G.B., Sobral, A., Marinho, M., Melo, A., Association between climate variables and global transmission oF SARS-CoV-2. Science of The Total Environment, 729 (2020). https://doi.org/10.1016/j.scitotenv.2020.138997
Huang, Z., Huang, J., Gu, Q., Du, P., Liang, H., Dong, Q., Optimal temperature zone for the dispersal of COVID-19. Science of The Total Environment, 729 (2020). https://doi.org/10.1016/j.scitotenv.2020.139487
Gardner, E., Kelton, D., Poljak, Z., Van Kerkhove, M., von Dobschuetz, S., Greer, A.L., A case-crossover analysis of the impact of weather on primary cases of Middle East respiratory syndrome. BMC Infectious Diseases, 19 (2019).https://doi.org/10.1186/s12879-019-3729-5
Tamerius, J.D., Shaman, J., Alonso, W.J., Bloom-Feshbach, K., Uejio, C.K., Comrie, A., Viboud, C., Environmental predictors of seasonal influenza epidemics across temperate and tropical climates. PLOS Pathogens, 9 (2013). doi:10.1371/journal.ppat.1003194.g001
Air pollution vs Virus Literature:
Zhu, Y., Xie, J., Huang, F., Cao, L., Association between short-term exposure to air pollution and COVID-19 infection: Evidence from China. Science of The Total Environment, 729 (2020). https://doi.org/10.1016/j.scitotenv.2020.138704
Liu, J., et al., Impact of meteorological factors on the COVID-19 transmission: A multi-city study in China. Science of The Total Environment, 726 (2020). https://doi.org/10.1016/j.scitotenv.2020.138513
Bashir, M.F., Ma, B., Bilal, Komal, B., Bashir, M.A., Tan, D., Bashir, M., Correlation between climate indicators and COVID-19 pandemic in New York, USA. Science of The Total Environment, 729 (2020). https://doi.org/10.1016/j.scitotenv.2020.138835
Wu, X., Nethery, R.C., Sabath, B., Braun, D., Dominici, F., Exposure to air pollution and COVID-19 mortality in the United States: A nationwide cross-sectional study. https://projects.iq.harvard.edu/covid-pm