OmopViewer

1. Overview
2. Key Features
3. Installation
4. Operational Architecture
5. Usage Examples
   1. 1. Generating a Static Deployable Shiny App
   2. 2. Launching the Dynamic Upload Viewer
6. Main Functions

The OmopViewer R package provides automated tools to build and deploy interactive R Shiny applications for visualizing, exploring, and sharing standardized study results formatted as <summarised_result> objects across the DARWIN EU and OHDSI ecosystems.

---

Overview

In multi-center observational research, sharing raw patient-level data is prohibited by data privacy regulations (such as GDPR and HIPAA). Instead, research networks export aggregated, privacy-preserving <summarised_result> data objects.

OmopViewer transforms these standardized summary results into interactive, web-based graphical dashboards without requiring manual Shiny coding. It provides two complementary operational modes:

1. Static Shiny App (exportStaticApp): Generates a self-contained, customizable Shiny project directory pre-populated with study results, ready for hosting on Shiny Server, Posit Connect, or shinyapps.io.
2. Dynamic Shiny App (launchDynamicApp): Launches an interactive dashboard where users can upload any results.csv or zipped bundle of summarised_result outputs and explore tables and interactive figures on the fly.

---

Key Features

• Standardized Result Compatibility: Native support for results from CohortCharacteristics, IncidencePrevalence, CohortSurvival, DrugUtilisation, MeasurementDiagnostics, PhenotypeR, and visOmopResults.
• Zero-Code Application Generation: Automatically creates UI, server, and preprocessing files for all recognized result types.
• Custom Theming & Branding: Supports custom organization logos, bslib Bootstrap themes, title customization, and Markdown landing pages (background.md).
• Granular Panel Navigation: Configurable panel hierarchies, dropdown menus, and result filters.
• Privacy Compliant: Visualizes only aggregated, de-identified summary estimates and suppresses small cell counts according to network minimum thresholds.

---

Installation

Install OmopViewer from CRAN or GitHub:

# From CRAN
install.packages("OmopViewer")

# Development version from GitHub
# install.packages("pak")
pak::pkg_install("ohdsi/OmopViewer")

---

Operational Architecture

graph TD
    subgraph "1. Upstream DARWIN EU Packages"
        A1["CohortCharacteristics"]
        A2["IncidencePrevalence"]
        A3["CohortSurvival"]
        A4["DrugUtilisation"]
        A5["MeasurementDiagnostics"]
    end

    subgraph "2. Standardised Result Format"
        R["summarised_result / results.csv"]
    end

    subgraph "3. OmopViewer Deployment"
        M1["exportStaticApp()<br/>Creates deployable standalone Shiny app"]
        M2["launchDynamicApp()<br/>Interactive drag-and-drop explorer"]
    end

    subgraph "4. Interactive Dashboard"
        D1["Demographics & Table 1"]
        D2["Incidence / Prevalence Trends"]
        D3["Survival Curves & Cox Models"]
        D4["Drug Persistence & Pathways"]
        D5["Measurement Distributions"]
    end

    A1 & A2 & A3 & A4 & A5 --> R
    R --> M1 & M2
    M1 & M2 --> D1 & D2 & D3 & D4 & D5

---

Usage Examples

1. Generating a Static Deployable Shiny App

Combine multiple study results and export a complete Shiny project:

library(CohortCharacteristics)
library(IncidencePrevalence)
library(OmopViewer)
library(omopgenerics)

# 0. Generate study results
cdm <- mockCohortCharacteristics()
char_res <- summariseCharacteristics(cdm$cohort1)
attr_res <- summariseCohortAttrition(cdm$cohort1)

# 1. Combine into a single result set
study_results <- bind(char_res, attr_res)

# 2. Export complete standalone Shiny app to a target directory
exportStaticApp(
  result = study_results,
  directory = "./study_shiny_dashboard",
  title = "Observational Study Results Dashboard",
  theme = "cerulean"
)

The exported directory contains:

• ui.R and server.R: Complete application logic.
• global.R: Package loading and initialization.
• data/result.csv: Privacy-preserving aggregated result table.
• data/studyData.RData: Preprocessed application data cache.

2. Launching the Dynamic Upload Viewer

Launch a local dashboard to inspect any study results file:

library(OmopViewer)

# Launch dynamic interactive explorer
launchDynamicApp()

---

Main Functions

Function	Purpose
exportStaticApp()	Generates a standalone, deployable Shiny application folder from one or more summarised_result objects.
launchDynamicApp()	Launches an interactive Shiny application in your browser allowing drag-and-drop upload and visualization of arbitrary OMOP study results.