A Shiny app for human reviewers checking and correcting AI-generated data extraction and Drummond quality-appraisal results against the original source PDFs, as part of an AI-assisted literature review of health economic evaluation studies. The source PDF and the AI’s extracted data sit side by side, so every field and QA verdict can be checked and corrected at a glance.
(An AI pipeline’s extraction/QA spreadsheets are split into per-reviewer batches with split_for_review() before reviewing starts, and combined back into the final verified deliverables with combine_reviewed() once it’s done – see vignette("CoReview") for details.)
Installation
You can install the development version of CoReview like so:
# (Only necessary if the remotes package is not installed).
utils::install.packages("remotes")
remotes::install_github("MUHEG/CoReview")Run
Launch the app with a folder picker, so a reviewer can load any batch folder themselves:
CoReview::run_app()Or point it straight at a batch, optionally pre-filling the reviewer name:
CoReview::run_app(batch_dir = "review_batches/batch_1", reviewer = "Jane Doe")Status
This library is experimental and under active development as part of an ongoing systematic review project. It is not currently recommended for use beyond members of the development team.
About
You are reading the doc about version : 0.0.0.9000
This README has been compiled on the
Sys.time()
#> [1] "2026-10-20 16:20:51 AEDT"Here are the tests results and package coverage:
devtools::check(quiet = TRUE)
#> ℹ Loading CoReview
#> ── R CMD check results ──────────────────────────────── CoReview 0.0.0.9000 ────
#> Duration: 1m 46.2s
#>
#> ❯ checking for future file timestamps ... NOTE
#> unable to verify current time
#>
#> ❯ checking top-level files ... NOTE
#> Non-standard file/directory found at top level:
#> 'Citation.cff'
#>
#> 0 errors ✔ | 0 warnings ✔ | 2 notes ✖
covr::package_coverage()
#> CoReview Coverage: 81.34%
#> R/run_app.R: 0.00%
#> R/mod_coreview_main.R: 54.55%
#> R/combine_reviewed.R: 94.74%
#> R/combine_logic.R: 95.24%
#> R/split_for_review.R: 98.82%
#> R/app_config.R: 100.00%
#> R/app_server.R: 100.00%
#> R/app_ui.R: 100.00%
#> R/extraction_definitions.R: 100.00%
#> R/golem_utils_server.R: 100.00%
#> R/golem_utils_ui.R: 100.00%
#> R/qa_definitions.R: 100.00%
#> R/review_io.R: 100.00%