Stan being used to study and fight coronavirus

Hi all. Someone was asking me where Stan has been used to study and fight coronavirus. I came up with a few examples; see below. If people have others, please share in this thread.

I’ve limited myself to examples where there is some report or public code; I did not include various ongoing projects that have not produced anything public yet.

Here’s a list so far:

This from the Imperial College team:

https://github.com/ImperialCollegeLondon/covid19model/tree/v1.0 and https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-Europe-estimates-and-NPI-impact-30-03-2020.pdf

This from University of Bern:

https://www.medrxiv.org/content/10.1101/2020.03.04.20031104v1.full.pdf (see discussion here: https://statmodeling.stat.columbia.edu/2020/03/09/coronavirus-model-update-background-assumptions-and-room-for-improvement/)

This from NYU Abu Dhabi:

https://osf.io/preprints/socarxiv/jp4wk and https://github.com/saudiwin/corona_tscs

This paper in New England Journal of Medicine:

https://www.nejm.org/doi/10.1056/NEJMc2004973 (see supplementary information here: https://www.nejm.org/doi/suppl/10.1056/NEJMc2004973/suppl_file/nejmc2004973_appendix.pdf)

11 Likes

@harrelfe’s RCT

http://hbiostat.org/proj/covid19/

4 Likes

https://scabras.github.io/cvirus/

3 Likes

Here’s a paper about N95 respirator decontamination from NIAID that uses Stan: https://www.medrxiv.org/content/10.1101/2020.04.11.20062018v1.full.pdf

Another one using Stan, on estimating the proportion of asymptomatic cases on the Diamond Princess: https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2020.25.10.2000180

4 Likes

Here’s another one on the importance of taking into account R0’s variability across regions and time through hierarchical modelling.

3 Likes

Came here to add my papers and they’re already here :) Thanks all!

1 Like

Though there’s a newer version of the N95 paper on OSF (waiting for it to clear on medRxiv): https://doi.org/10.31219/osf.io/phcsb

4 Likes

Risk for Transportation of 2019 Novel Coronavirus Disease from Wuhan to Other Cities in China

3 Likes

Analysis on the trend of the growth rate and the detection rate on Japan and Korea
https://github.com/yoriyuki/COVID-19/blob/master/prediction/Turzin.ipynb (Japan)
https://github.com/yoriyuki/COVID-19/blob/master/prediction/Turzin-Korea.ipynb (Korea)
No paper yet. Some explanation in Japanese
https://note.com/yoriyuki/n/n1fc028da0b68
I’m still improving a model and also want to do model selection (and validation).

3 Likes

Someone pointed me to this one from the University of Texas COVID-19 Modeling Consortium: https://covid-19.tacc.utexas.edu/projections/ with writeup here: https://covid-19.tacc.utexas.edu/media/filer_public/87/63/87635a46-b060-4b5b-a3a5-1b31ab8e0bc6/ut_covid-19_mortality_forecasting_model_latest.pdf

3 Likes
4 Likes

A paper by Simas Kucinskas, Tracking R of COVID-19: (https://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=2441138)

3 Likes

Update. I write a paper.

https://github.com/yoriyuki/BayesianCOVID19/blob/master/paper/BayesianCOVID-19.pdf

2 Likes

The paper is in the works, but I’m in Uni Iceland’s modelling group and our predictions for Iceland were very good. We have been surprised in how well we predicted total cases, and using percentages from Imperial College’s paper from March 16 we were also surprisingly spot on with hospital and ICU predictions.

We’ve been updating our website steadily at: https://covid.hi.is/english/

Technical writeup about the modelling at: https://rpubs.com/bgautijonsson/HierarchicalLogisticGrowthCurves

One good thing that’s come of our website is the Icelandic news has used our figures without having to make their own and having some information lost or accidentally changed in the process.

6 Likes

Some models for errors in prevalence studies:
https://statmodeling.stat.columbia.edu/2020/05/01/simple-bayesian-analysis-inference-of-coronavirus-infection-rate-from-the-stanford-study-in-santa-clara-county/
and
https://statmodeling.stat.columbia.edu/2020/05/04/bayesian-analysis-of-santa-clara-study-run-it-yourself-in-google-collab-play-around-with-the-model-etc/
and

2 Likes

I don’t know if this thread is still being used, but here’s a new one: Bayesian adjustment for preferential testing in estimating the COVID-19 infection fatality rate: Theory and methods.

4 Likes

Yes, please continue adding to the thread. Thank you.

Another:

Community prevalence of SARS-CoV-2 in England: Results from the ONS Coronavirus Infection Survey Pilot

1 Like