search

neherlab / covid19_scenarios_data

Data preprocessing scripts and preprocessed data storage for COVID-19 Scenarios project
https://github.com/neherlab/covid19_scenarios
Other
41 stars 36 forks source link

🇳🇱 feat: add icu data for netherlands #35

Closed hongaar closed 4 years ago

hongaar commented 4 years ago

Added data source for ICU cases for The Netherlands from Stichting NICE: https://www.stichting-nice.nl/. They don't mention any license for the data, this is the description from their website (translated with Google):

Since the COVID-19 outbreak at the end of February, the Dutch ICs have registered these in the NICE registration. These data will be linked at a later stage to the regular NICE data regarding, among other things, disease severity, demographics, mortality and treatment duration. Below is an overview of the number of (current) recordings with a COVID-19 infection on the Dutch ICs. For the compilation of this overview, we depend on data delivery by the participating ICs. Due to pressure on the IC, there may be a delay of 2 to 3 days in the data delivery.

My knowledge of Python is very limited. Open to hear how I can improve the script.

hongaar commented 4 years ago

I can also look into adding Dutch provinces based on https://github.com/J535D165/CoronaWatchNL.

PS: They have a CI job setup to download and parse data every hour: https://github.com/J535D165/CoronaWatchNL/blob/master/.github/workflows/download.yml I can adapt it for this repo if that's useful?

nnoll commented 4 years ago

@hongaar: This looks great -- thank you so much! Would you mind filling out an entry for the Netherlands within populationData.tsv? Without this, the case counts you have kindly provided us won't be loaded as a seperate scenario. Once I have this I'll accept the pull request.

Regarding your second point -- if you can provide provinces that would be great! And we have talked about adding a bot that can pull new data if it's available for this repo for a bit now. If you want to have a go at it, that would be awesome!

rneher commented 4 years ago

We would ideally have cases, deaths, and ICU in one file. The file you currently output only writes ICU patients to file. Is there a comparable source for the other fields? Otherwise this could be merged with the data aggregated by ECDC

hongaar commented 4 years ago

We would ideally have cases, deaths, and ICU in one file. The file you currently output only writes ICU patients to file. Is there a comparable source for the other fields? Otherwise this could be merged with the data aggregated by ECDC

Yeah I was not sure how to approach this. The source I added only has ICU data. Does it make sense to pull in more (duplicate?) data into the .tsv file for The Netherlands, or does the parse_all script merge it already?

rneher commented 4 years ago

We currently forsee only one .tsv file per location. So it would be ideal if we could merge this before generating the .tsv and then funnel all of this into the json.

hongaar commented 4 years ago

@rneher added separate sources for cases, deaths and hospitalized to .tsv.

@nnoll there was already an entry in populationData.tsv, should I add anything else? Created #47 for the scheduled job.