DetectorCheckerWebApp

Created by: Julia Brettschneider (original R code), Tomas Lazauskas (R package engineering), Oscar Giles (package development) and Wilfrid Kendall (testing and editing).

Overview

An interactive WebApp for analysing pixel damage in CT scanners using the DetectorChecker R package (https://github.com/alan-turing-institute/DetectorChecker). While the target application concerns CT scanners, this application can also be used to analyze screen damage arising from other sources.

This WebApp loads data acquired from detector screens (tif images from daily tests, or xml files listing bad pixels), and applies methods from spatial statistics (particularly the R package spatstat) to produce analyses and graphical reports concerning location and possible departures from spatial randomness of damaged pixels and regions.

Running the app

Official release

The official release of the DetectorChecker WebApp is hosted at https://detectorchecker.azurewebsites.net.

Documentation

A user guide is available as a pdf file.

Running the app from source

There are several ways to launch the WebApp from source, which is strictly for development purposes. First clone the repository.

Make sure you have the latest version of the DetectorChecker R package installed (see https://github.com/alan-turing-institute/DetectorChecker).

Dependencies

To run the WebApp from source make sure you have the following R packages installed:

install.packages(c("shiny", "shinyjs", "shinyBS", "shinythemes", "spatstat", "shinydashboard"))

Run with R (limited functionality)

1. Navigate to the webapp subdirectory and run the run.R file.

2. Open a browser and go to http://0.0.0.0:1111

Shiny App (running locally)

The WebApp can be run as a shiny app simply by running the bash script in the main directory:

1. Run the bash script from the terminal:

bash shiny.sh

2. Open a browser and go to http://0.0.0.0:1111

Limited functionality: The WebApp will run, but you will not be able to upload data to the cloud.

Docker image

It is possible to run the WebApp as a Docker image (https://www.docker.com). See the instructions on how to build the Docker image.

Repository structure

The repository contains two sub folders:

docs: Documentation with a description of how to build the WebApp for developers.

webapp: R and Python scripts.

Data management (Azure)

Naming convention

email@address/yyyy-mm-dd-hh-mm-ssssss/layout_name.ext
email@address/yyyy-mm-dd-hh-mm-ssssss/layout_name.dc

Citation

If you use DetectorChecker in your work please cite our package.

BibTeX:

  @Misc{,
    title = {{DetectorChecker}: Assessment of damage to CT scanners},
    author = {Tomas Lazauskas and Julia Brettschneider and Oscar Giles and Wilfrid Kendall},
    url = {https://github.com/alan-turing-institute/DetectorChecker},
  }

Getting help

If you found a bug or need support, please submit an issue here.

How to contribute

We welcome contributions! If you are willing to propose new features or have bug fixes to contribute, please submit a pull request here.