EndTB

The goal of EndTB is to provide a convenient set of the established TB models for the simulation, estimation, and projection of Tuberculosis Epidemic.

Moreover

  • the models are implemented in C++ which provides fast simulations.
  • the models are called within the TMB framework, allowing advanced and fast parameter estimation with emprical Bayes approach.

Installation

You can install the development version of EndTB like so:

remotes::install_github('kklot/EndTB')

Documents of the package and functions can be view as normal R’s documentation or online at EndTB.

Example

This is a basic example which shows you how to start a model:

library(EndTB)
#> Welcome to EndTB
TB <- TBM$new(c(sigma = 2))

where the default parameters σ was replaced with 2.

Steady-state of the model without treatments and no population growth for that set of parameters can be ploted with.

TB$plot()

List of states to be plotted can be supplied into the arguments of the plot function above.

Further details can be read in References in the top navigation. For example, the documentation of plot.

Check out model’s parameters and states for Moldova model.

EndTB:::states_moldova
#>                        S                      LS1                      AS1 
#>            "Susceptible"         "Sen-1st-Latent"   "Sen-1st-Asymptomatic" 
#>                      IS1                      DS1                      ES1 
#>    "Sen-1st-Symptomatic"      "Sen-1st-Presented"       "Sen-1st-Seekcare" 
#>                      FS1                      LS2                      AS2 
#>          "Sen-1st-Treat"         "Sen-2nd-Latent"   "Sen-2nd-Asymptomatic" 
#>                      IS2                      DS2                      ES2 
#>    "Sen-2nd-Symptomatic"      "Sen-2nd-Presented"       "Sen-2nd-Seekcare" 
#>                      FS2                      RS0                      RS1 
#>          "Sen-2nd-Treat" "Sen-2nd-Recover-Stable"    "Sen-2nd-Relapse-Low" 
#>                      RS2                      LR1                      AR1 
#>   "Sen-2nd-Relapse-High"         "Res-1st-Latent"   "Res-1st-Asymptomatic" 
#>                      IR1                      DR1                      ER1 
#>    "Res-1st-Symptomatic"      "Res-1st-Presented"       "Res-1st-Seekcare" 
#>                      FR1                      LR2                      AR2 
#>          "Res-1st-Treat"         "Res-2nd-Latent"   "Res-2nd-Asymptomatic" 
#>                      IR2                      DR2                      ER2 
#>    "Res-2nd-Symptomatic"      "Res-2nd-Presented"       "Res-2nd-Seekcare" 
#>                      FR2                      RR0                      RR1 
#>          "Res-2nd-Treat" "Res-2nd-Recover-Stable"    "Res-2nd-Relapse-Low" 
#>                      RR2 
#>   "Res-2nd-Relapse-High"
EndTB:::pars_moldova
#>           N      beta_s      beta_r       kappa           b          mu 
#> 2.38428e+07 4.40000e+00 1.20000e+01 1.00000e-01 1.00000e-02 1.50000e-02 
#>       mu_tb     theta_s     theta_r         rho       sigma       delta 
#> 1.50000e-01 1.40000e-01 1.40000e-01 1.00000e-03 7.90000e+00 7.50000e+00 
#>       gamma         phi  varepsilon       omega       tau_0       tau_1 
#> 1.20000e+01 5.20000e+01 9.30000e-01 9.30000e-01 2.00000e+00 5.00000e-01 
#>       chi_s       chi_r      varrho         r_0         r_1         r_2 
#> 2.00000e-01 2.50000e-01 5.00000e-02 3.20000e-02 1.40000e-01 1.50000e-03 
#>         r_3    varsigma        c_s0        c_r0        c_r1         m_n 
#> 7.00000e-01 5.00000e-01 8.00000e-01 3.00000e-01 4.80000e-01 7.40000e-01 
#>         m_r          xi 
#> 4.70000e-01 8.00000e-01

TODO

  • 1st revision of the WHO’s model (and currently the only)
  • Revise model to Argentina data
  • Add fitting example
  • Add other model variations