OHDSI / OmopViewer

https://ohdsi.github.io/OmopViewer/
Apache License 2.0
2 stars 0 forks source link

OmopViewer

R-CMD-check CRAN
status Lifecycle:
experimental Codecov test
coverage

The goal of OmopViewer is to allow the user to easily create Shiny Apps to visualise study results in <summarised_result> format.

Installation

Install it from cran:

install.packages("OmopViewer")

Or you can install the development version of OmopViewer from GitHub with:

install.packages("pak")
pak::pkg_install("OHDSI/OmopViewer")

Main functionalities

library(OmopViewer)

The package has two functionalities:

Static shiny app

The static shiny app functionality creates a static shiny from a list of summarised_result objects. This shiny is specific to the set of results and can be modified later locally.

# lets generate some results
library(CohortCharacteristics)
#> Registered S3 method overwritten by 'visOmopResults':
#>   method                 from        
#>   tidy.summarised_result omopgenerics
cdm <- mockCohortCharacteristics()
#> Note: method with signature 'DBIConnection#Id' chosen for function 'dbExistsTable',
#>  target signature 'duckdb_connection#Id'.
#>  "duckdb_connection#ANY" would also be valid
result <- summariseCharacteristics(cdm$cohort1) |>
  bind(summariseCohortAttrition(cdm$cohort1))
#> ℹ adding demographics columns
#> ℹ summarising data
#> ✔ summariseCharacteristics finished!
#> `cohort_definition_id` casted to character.

exportStaticApp(result = result, directory = tempdir())
#> ℹ Processing data
#> ✔ Data processed: 2 panels idenfied: `summarise_characteristics` and
#>   `summarise_cohort_attrition`.
#> ℹ Creating shiny from provided data
#> `cohort_definition_id` eliminated from settings as all elements are NA.
#> ✔ Shiny created in:
#>   /var/folders/pl/k11lm9710hlgl02nvzx4z9wr0000gp/T//RtmpYP9P88/shiny

This function allow some customisation of the shiny with the arguments:

The shiny generated will have the following structure:

Dynamic shiny app

The dynamic shiny app can be easily launched with launchDynamicApp() function. This function creates a shinyApp where you can upload multiple results sets and visualise them.

launchDynamicApp()

By default the shiny generated will have no data, you have to upload data from a csv or zip file that you have it locally. The summarised_results will be processed and you will be allowed to choose which results to visualise.