Installation Instructions

PACSIFIER can be installed in several ways, ordered from simplest to most flexible:

Option 1: Pre-built Docker Image (Recommended) — Pre-built Docker image from Quay.io (recommended for most users)
Option 2: Install via pip — Install via pip from PyPI
Option 3: Build Docker Image Locally — Build the Docker image locally
Option 4: Install from Source (For Developers) — Install from source (for developers)

Option 1: Pre-built Docker Image (Recommended)

The simplest way to use PACSIFIER. No need to clone the repository or install any dependencies — everything is bundled in the Docker image.

Prerequisites

Ensure Docker is installed on your system:

Linux: Follow the official Docker installation guide.
Windows (Recommended: WSL): Install Docker Desktop for Windows and enable WSL 2 integration. Follow the Docker Desktop installation guide.
macOS: Follow the Docker Desktop installation guide.

On Linux, set Docker to be managed as a non-root user:

sudo groupadd docker
sudo usermod -G docker -a $USER
# Reboot, then verify:
docker run hello-world

Pull and Run

Pull the latest release image from Quay.io:

docker pull quay.io/translationalml/pacsifier:latest

Test that it works:

docker run --rm quay.io/translationalml/pacsifier:latest pacsifier --version

Run a query:

docker run --rm --net=host \
    -v /path/to/my_dir:/base \
    quay.io/translationalml/pacsifier:latest \
    pacsifier --save --info --queryfile /base/my_query.csv \
        --config /base/my_config.json --out_directory /base/my_output_dir

Tip

You can pin to a specific version instead of latest (e.g., quay.io/translationalml/pacsifier:1.0.0). See available tags on Quay.io.

Note

DCMTK and all other dependencies are included inside the Docker image. You do not need to install them separately.

We refer to Running PACSIFIER commands in Docker for more details on running the Docker image, and Docker Wrappers for convenience wrapper scripts that simplify the docker run command syntax.

Option 2: Install via pip

PACSIFIER is published on PyPI:

pip install pacsifier

Important

When installing via pip, you must also install DCMTK on your system for DICOM network communication:

Linux (Ubuntu-based): sudo apt install dcmtk
Windows (WSL): Same command within your WSL environment
macOS: brew install dcmtk

For more details, see the DCMTK documentation.

Verify the installation:

pacsifier --help

Option 3: Build Docker Image Locally

If you need the latest development version or want to customize the Docker image, you can build it from the repository:

git clone https://github.com/TranslationalML/pacsifier.git
cd pacsifier
make build-docker

Inspect the built image:

docker images | grep pacsifier

Test the image:

docker run --rm pacsifier:latest pacsifier --version

Note

DCMTK is included inside the Docker image. You do not need to install it separately.

Option 4: Install from Source (For Developers)

For developing or customizing PACSIFIER:

Install DCMTK (required for DICOM network communication):
```
sudo apt install dcmtk
```
For other operating systems, see the DCMTK documentation.

Clone the repository:

git clone https://github.com/TranslationalML/pacsifier.git
cd pacsifier

Create a Python environment (Python >= 3.10 required):

Using conda with a manual environment:

conda create -n pacsifier_minimal python=3.10
conda activate pacsifier_minimal

Or using the provided environment file:

conda env create -f environment/environment_minimal_202401.yml
conda activate pacsifier_minimal

Or using venv:

python3 -m venv venv
source venv/bin/activate

Install PACSIFIER in editable mode:
```
pip install -e .
```
For full development (documentation, tests, linting):
```
pip install -e ".[all]"
```
Verify the installation:
```
pacsifier --help
```
This should display the help message for PACSIFIER, confirming the installation was successful.