Skip to content

miraisolutions/Covid19

Repository files navigation

Covid19Mirai

R build status eRum2020::CovidR

The goal of Covid19Mirai is to provide an insight on corona virus data taken from public and reliable resources.

Data Source history

Up to the 26th of March 2020, the source was the Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE)

However, due to a format change on the data source, we had to switch, starting the 26th of March 2020, to the work of David Bumbeishvili, who is maintaining the old JHU CSSE data set format with updates declared as from worldometers. This is a new project and we cannot guarantee the long-term reliability of the data source. We are, however, grateful to David Bumbeishvili for his work.

Update 4th of July 2020, Data from few European countries have been readjusted recently, getting this update has given us the push to switch to a new and richer data source. We have decided for the COVID 19 Data Hub project led by Emanuele Guidotti and David Ardia. We are very thankful to David Bumbeishvili for his great work

Update October 2023, the app stops updating data and is frozen to 2022-09-01.

Data Storage

Data are updated with a delay of 40h, i.e. at 4pm CEST the Last date is updated taking the yesterday date. This allows having data for about all countries when the date is updated with the new one.

The data are stored as RDS file in folder inst/datahub of the package. A script build_data runs in GitHub Actions twice every day to update the data in the package.

The data update was stopped in October 2023.

The dashboard

It consists of 5 main pages:

  • Global: Summarizing top 5 countries in the world in each variable.
  • Continents: Summarizing world data per continent (defined according to UN), underneath sub-tabs with insight within continent:
    • Europe: European data split by macro-areas with heat-maps per country.
    • Asia: Asia data split by macro-areas with heat-maps per country.
    • Africa: African data split by macro-areas with heat-maps per country.
    • Latin America & Carib.: South and Central American with Caribbean Isles by macro-areas with heat-maps per country.
    • Northern America: Northern America (USA and Canada) data split by macro-areas with heat-maps per country.
  • Switzerland: Single country report of Switzerland.
    • Maps and graphs of Cantonal data displayed.
  • Country: Single country report.
    • If available regional graphs and data will be displayed.
  • Country Comparison: Comparison report between N countries from all over the world.

Macro areas of Continents are defined following United Nations indications.

The Variables

The Covid19datahub project can allow us to use the following variables:

  • confirmed: number of confirmed cases. Usually tested positive.
  • recovered: number of healed or tested negative cases. Some countries have stopped reporting recovered cases.
  • deaths: number of dead confirmed cases.
  • tests: number of tests. Not available for all countries.
  • hosp: number of currently hospitalised confirmed cases. Not available for all countries.
  • icuvent: number of currently hospitalised Ventilated or in Intensive Care. Not available for all countries. Categorization differed from countries to country therefore Ventilated and Intensive Care variables have been aggregated.
  • stringency index: Lock Down stringency index from 0 to 100.
  • vaccines: number of vaccine doses given to the population.
  • population: population size.

The following variables are computed by the application:

  • active: number of active cases, usually tested positive. confirmed - recovered - deaths.
  • prevalence over 1M: number of confirmed cases per 1 Million inhabitants.
  • growth factors (3 7 14): number of confirmed cases today divided by the number of confirmed cases 3 7 14 days ago over the previous 2 months.
  • lethality rate: number of deaths divided by the number of confirmed cases.
  • mortality rate: number of deaths divided by the population size.
  • positive test rate: number of positive tests, i.e. confirmed divided by the tests carried in the day.
  • int care hospitalised rate: number of patients currently in Intensive Care / Ventilated status, divided by the number of currently hospitalised patients.
  • vaccines rate: number of vaccines divided by the population size.
  • new (variable): all variables labelled “new” are the delta of day X value - day X-1.
  • last week (variable): all variables labelled “last week” are the totals of the last 7 days.
  • past week (variable): all variables labelled “past week” are the totals of the previous 14 days to the last 7 days.
  • last month (variable): all variables labelled “last month” are the totals of the last 30 days.

The results are visualized as a shiny app.

Installation

You can install Covid19Mirai from GitHub with:

# install.packages("devtools")
devtools::install_github("miraisolutions/Covid19")