Skip to content

ThomUK/PathSimR

 
 

Repository files navigation

PathSimR

This is a fork of a project from https://github.com/nhs-bnssg-analytics/PathSimR

There is a 15 minute video about how to use the tool (made by the BNSSG team) available here: https://vimeo.com/643579424

This fork aims to refactor & modularise the {shiny} code, better understand how the app works, and potentially convert it in future to use {simmer}. At the point of forking the app was a single R file over approx 14K lines long, which I found difficult to study.

Installation

You can install PathSimR like so:

# install.packages("remotes")
remotes::install_github("ThomUK/PathSimR")

To run the app on your local machine, use the below:

library("PathSimR")
run_app()

or simply:

PathSimR::run_app()

Using the tool

For a quick overview see the video (made by the BNSSG team) available here: https://vimeo.com/643579424

Broadly, using the tool involves following the instructions on the screen. There is also documentation available within the PathSimR_Shiny directory. This includes the folders below:

  1. Documentation contains the technical documentation for the app and a sample use case template to capture modelling requirements.

  2. Use Case Library contains expamples of requirements, inputs, and outputs for four case studies, as an illustration of how the tool can be used

  3. Network Templates contains sample input templates which can be loaded into the tool (as an alternative to using the inbuilt input wizard). They are intended to demonstrate features of the input format and not to represent real pathways. Files in this format can be downloaded, modified, and re-uploaded to PathSimR to save time, after an initial pathway has been created using the input wizard. A simple example of pair of input templates which will create an error are also included in this folder, as an aid to identifying problem inputs.

Report template files

The PathSimR_Report.Rmd and the template.docx files are called from within the app to create a downloadable Rmarkdown report and do not need to be opened or run separately unless you wish to alter the report format (e.g. replace the Word template with one containing your own organisation’s branding).

About

R-based discrete event simulation for modelling patient pathways. Sponsored by The Health Foundation.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 100.0%