There is a lot of data available about Covid-19, a lot of research is being done focused on the number of cases, trying to predict deaths and ways to reduce the number of cases. Many other factors are not taken into account, including socioeconomic factors like HDI, average income, life expectancy, and data related to environmental events. Can these be relevant when deciding what to do against the virus? Our proposal aims to connect and find correlations that are useful and help to compare different regions. We will unite two worlds that seem totally different and give cities efficient tools to fight Covid-19.
What inspired us is seeing the horrific times we’re going through and having the urge to do something to help by giving some previously unthought insights. We take the attitude of helping, with all our potential and available time, using the climatic data from NASA, socioeconomic data from the IBGE (Brazilian Institute of Geography and Statistics) and publicly available Covid related data. The tools we use:
The problems we had:
Our achievements:
Here is a link to our web application: https://viridi19.netlify.app/
NOAA - Climate.gov
- https://climate.gov/maps-data/primer/what-environmental-data-are-relevant-study-infectious-diseases-covid-19
We used data from NOAA studies to support our analyzes regarding the use of environmental variables in the process of detecting correlations between the spread of the virus.
Meteomatics API
- https://www.meteomatics.com/en/api/overview/
We used data from one of the challenge partners to extract climatic variables from cities in Brazil, they were used for anomaly detection analysis using time series averaging techniques.
National Institute of Meteorology - (INMET - National Institute of Meteorology)
- http://www.inmet.gov.br/portal/index.php?r=estacoes/estacoesConvenciones
We used data made available by government agencies in Brazil, data extracted from meteorological satellites, with granularity of days. This data set was essential to perform analyzes and eliminate possible noise from other extracted databases. For the extraction of features, we use scientific studies ([1], [2], [3], [4], [5]) that analyzed climatic impacts and that resulted in a set of significant features that may be related to proliferation of the virus.
Atlas of human development in Brazil
- https://www.br.undp.org/content/brazil/pt/home/idh0/rankings/idhm-municipios-2010
We extracted social and economic development metrics in Brazil, referring to the last update. This extraction took place for all municipalities in the country. Containing several features that represent country characteristics, such as:
Economic development index, health development indexes, educational development indexes, number of the population divided by different age groups, number of population in different educational development ranges and many other features that represent possible social inequalities in the municipality in question.
These data were extracted in order to perform feature engineering and create methods and models to generate an algorithm that calculates the similarity between cities based on these data. Initially we used cosine distance (https://en.wikipedia.org/wiki/Cosine_similarity) between all cities, extracting an index that represents how close those data are, from there, we were able to make a ranking of cities based on a search.
Data from covid-19 do Brasil
- https://covid19br.wcota.me
We used data from covid-19 in brazil, with granularity of days for each municipality in the country, with the intention of making the necessary comparisons between the contamination curves.
References:
[1] Ahmadi, M., Sharifi, A., Dorosti, S., Jafarzadeh Ghoushchi, S., & Ghanbari, N. (2020). Investigation of effective climatology parameters on COVID-19 outbreak in Iran. Science of The Total Environment, 729, 138705. https://doi.org/https://doi.org/10.1016/j.scitotenv.2020.138705
[2] Alvarez-Ramirez, J., & MERAZ, M. (2020). Role of meteorological temperature and relative humidity in the January-February 2020 propagation of 2019-nCoV in Wuhan, China. MedRxiv, 2020.03.19.20039164. https://doi.org/10.1101/2020.03.19.20039164
[3] Auler, A. C., Cássaro, F. A. M., da Silva, V. O., & Pires, L. F. (2020). Evidence that high temperatures and intermediate relative humidity might favor the spread of COVID-19 in tropical climate: A case study for the most affected Brazilian cities. Science of The Total Environment, 139090. https://doi.org/https://doi.org/10.1016/j.scitotenv.2020.139090
[4] Baker, R. E., Yang, W., Vecchi, G. A., Metcalf, C. J. E., & Grenfell, B. T. (2020). Susceptible supply limits the role of climate in the COVID-19 pandemic. MedRxiv, 2020.04.03.20052787. https://doi.org/10.1101/2020.04.03.20052787
[5] Berumen, J., Schmulson, M., Guerrero, G., Barrera, E., Larriva-Sahd, J., Olaiz, G., … Tapia-Conyer, R. (2020). Trends of SARS-Cov-2 infection in 67 countries: Role of climate zone, temperature, humidity and curve behavior of cumulative frequency on duplication time. MedRxiv, 2020.04.18.20070920. https://doi.org/10.1101/2020.04.18.20070920