Install
The desktop app is the recommended dicomqc interface. The CLI supports automation and scripted workflows; both use the same Python audit engine.
Desktop App
Installers have not been published. To run the app now, follow the desktop source-build instructions. You need Python, Node.js, Rust, and the operating system's Tauri prerequisites.
Once it is running, try the Desktop privacy audit example.
CLI
Install from PyPI with Python 3.10 or newer. The PyPI package installs the CLI, not the desktop app.
Install from PyPI
Create an isolated environment so dicomqc and its dependencies do not alter the system Python installation:
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install dicomqc
Verify the installed command and version:
dicomqc --version
Upgrade an existing installation with:
python -m pip install --upgrade dicomqc
Optional MultiQC installation
dicomqc writes standalone HTML reports without extra packages. Install MultiQC only if you want to combine dicomqc results with other QC reports:
python -m pip install multiqc
See the scan walkthrough for the rendering command and its matching report screenshots.
Install from source
From a repository checkout, install dicomqc and its runtime dependencies in editable mode:
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e .
Contributors who need pytest and coverage tooling can install the test
optional dependency group:
python -m pip install -e ".[test]"
The test extra is not needed for normal use.
For the complete suite, including the optional local API, install
".[test,api,api-test]". The desktop app has separate native build
requirements; installing the PyPI package provides the CLI.
External DICOM tools
DCMTK, Orthanc, and other pseudonymization or remediation tools are not dicomqc dependencies. Install them separately only when they are part of the local DICOM transformation workflow. dicomqc itself remains read-only.
Continue with the Desktop example or CLI walkthrough to run your first audit with synthetic data.