micb25 / corona-jena

Erfassung und Visualisierung der Corona-Fallzahlen für Thüringen.

Hi there, I’m Michael 👋

I’m Michael Böhme, theoretical chemist 👨🏻‍🔬 and computer enthusiast 👨🏻‍💻 from Jena 💚, Germany. I’m currently working part-time at the Friedrich Schiller University in Jena doing scientific research in computational magnetochemistry. 🧲 You can have a look on my personal website, if you want to know more about my scientific work. 🧪

Besides, I’ve loved to code since I was 10 years old. 👦🏻 I learned to program on an old Commodore C64 👾 with the help of the great C64 User's Guide. 📖 Nowadays, I am mostly working with C++ and Python, sometimes also MatLab, BASH scripts or gnuplot. 📈 I’m currently learning a lot of 3D graphics programming in Vulkan. 🌋 In my spare time 🌞, I’m working on a yet secret project of mine. 😎 Stay tuned! 🖖

corona-jena's Issues

integration of "intensivregister" and especially the dynamic data development in the register

may be that this suggestion exceeds the project target. But as an evident part of the covid19 data record it would be nice to integrate the "intensivregister" and especially the dynamic data development in the register.
e.g, in thuringia is to see: some healthcare stakeholders dont bring the data also 2 weeks after the obligation "to bring".

The law "obligation to bring" in thuringia:
https://www.tmasgff.de/fileadmin/user_upload/Gesundheit/Dateien/COVID-19/3._ThuerSARS-CoV-2-EindmassnVO.pdf i.V.m.
https://www.tmasgff.de/fileadmin/user_upload/Gesundheit/Dateien/COVID-19/20200407_Versorgungskonzept.pdf i.V.m.
https://www.thueringen.de/mam/th7/tmsfg/gesundheit/krankenhauswesen/krankenhausplan_thueringen.pdf

Some comments to actual situation of implementation by obligation "to bring":
-> https://demenzpflaster.wordpress.com/2020/04/20/betteninventur-thueringen-17-04/
-> https://demenzpflaster.wordpress.com/2020/04/13/divi-intensivregister-betteninventur-in-thueringen/

The origin INTENSIVREGISTER:
https://www.intensivregister.de/#/intensivregister

The daily updated JSON - API of INTENSIVREGISTER (state of Apr. 28th.):
https://interaktiv.morgenpost.de/corona-deutschland-intensiv-betten-monitor-krankenhaus-auslastung/data/intensivregister.json
https://www.intensivregister.de/api/public/intensivregister (raw JSON)

2 examples for parsing the API data (after dl):
jq '.data[] | select(.krankenhausStandort.bundesland == "THUERINGEN") | .meldezeitpunkt + " co= " + ( .faelleCovidAktuell | tostring ) + " " + .label + " HighCare: " + .bettenStatus.statusHighCare'
jq '.data[] | select(.faelleCovidAktuell >= 10) | ( .faelleCovidAktuell | tostring ) + " ECMO: " + .bettenStatus.statusECMO + " " + .label'

example of result "select(.faelleCovidAktuell >= 20)"
http://ix.io/2k6p

the universe of INTENSIVREGISTER on GitHub:
https://github.com/search?q=intensivregister

Other interesting question:
following the INTENSIVREGISTER hospital names and postal adresses:
-> investigation of the number of beds and also more parameters by scraping some crawled sites like the homepages of the hospitals or some existing local overwiew sides where sometimes the numbers of beds are hided.

use CSV format for data storage

CSV should be used as a primary file format for data storage. This needs the crawlers as well as most of the gnuplot files to be rewritten.

How about using Chart.js?

Maybe you want to consider using Chart.js to display the graph-data.
In my opinion it is more user-friendly, HTML5 and there are some features like cursors.

Check the repository

If you want some help, let me know.

Interactive map of Thuringia

1. JSON
   Add possibility to obtain the following JSON structure for the interactive map. For now the map generates its data using the CSV. Maybe just generate a file containing this string (can be minified - without spaces) everytime you crawl for new data.

json_structure.txt

2. Publish Map in Linklist or include in mainsite to replace static maps

incorrect data entry for Jena, 2020-12-31 23:59

The data for Jena, 2020-12-31 23:59 was entered as 2021-12-31 23:59. Those data are actually not yet inside the repo, but will be fetched using data/update_cases.sh (which is inside the repo) and produces a rather wrong https://michael-böhme.de/corona/plot3_Jena.png.

Use active cases to compute doubling times

At the moment we are using all cases, to compute the doubling times, but now that we are nearly done with the first wave (at least for Jena) these numbers might be misleading, because a high infectious activity might be hidden in the old cases.
Wouldn't it be informative to compute the doubling times from the active cases?

add crawler for RKI case numbers for Thuringia

A crawler for the RKI case numbers for Thuringia needs to be implemented.

Switch source of count for Thuringia

Maybe you should use the site from the R.Koch institute to get the values for Thuringia - https://experience.arcgis.com/experience/478220a4c454480e823b17327b2bf1d4
Maybe more solid and easier to parse them.

best regards.