Skip to content

timueh/PandemicModeling

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Modeling pandemics subject to stochastic uncertainties -- A polynomial chaos approach

The so-called SEIR model is commonly used to model the outbreak of pandemics. The plain SEIR model consists of four differential equations, modeling the dynamics of susceptible, exposed, infected, and recovered people. Undeniably, a portion of infected people needs intensive care. To account for this fact variants of the model exist, see for instance here for the current Covid-19 pandemic (German), or here (German), or here (English). As with every mathematical model there are parameters that need to be chosen. Usually, not precise figures exist. Hence, uncertainty quantification can play a vital role.

General resources (mostly Germany-related)

Resource Explanation
corona.rki.de Dashboard by Robert Koch Institut; goes down to communal level
corona.rki.de -- API API for resource from above
RKI - Meldedaten Meldedaten nach Infektionsschutzgesetz, Robert Koch Institut
Epidemiologisches Bulletin by Robert Koch Institut
Epidemic calculator Epidemilogical simulation tool
CovidSim Epidemilogical simulation tool (for Europe)
Risklayer Explorer This company grew out ouf KIT; they still work together closely. Provides more than just information about Germany.
The Mathematics of Infectious Diseases Great SIAM paper by Herbert W. Hethcote
Data against Covid Initiative to share and provide expertise
Estimating $R_0$ Estimation of basic reproduction number for Germany
SIAM Resources for Covid-19 Excellent collection of online resources

Results

The code introduces uncertainty with respect to the basic reproduction number and the percentage of infected persons that require intensive care (see the code for precise numbers). Unfortunately, the numbers sketch a rather glim future (here for the German scenario). For instance, here is the number of infected patients over the course of one year.

Number of infected patients

And here the number of patients required intensive care

Number of infected patients

Model

The basic equations for the SEIR model are (see for example this excellent reference)

Equations for SEIR model

and we augment them by equations for patients requiring intensive care units (ICUs) motivated by this press release

Equations for ICU model

Code

The code is written in Julia. To run it, install all required packages. The main dependencies are DifferenialEquations.jl and PolyChaos.jl.

Documentation

This repository contains Julia code and a documentation to show how polynomial chaos expansion can help quantify uncertainties for the SEIR model.

Releases

No releases published

Packages

No packages published

Languages