Obtaining and running the Gadgetron

We recommend two different ways of obtaining and running the Gadgetron; using a conda environment or using Docker containers.

Installing in conda environment

The Gadgetron can be installed in a conda environment. To install the Gadgetron define and environment.yml file with:

name: gadgetron
channels:
  - ismrmrd
  - gadgetron
  - nvidia/label/cuda-11.6.1
  - conda-forge
  - bioconda
  - defaults
  - intel
dependencies:
  - gadgetron>=4.4.3
  - siemens_to_ismrmrd>=1.2.6

And create the environment with:

conda env create -f environment.yml

After activating the environment (with conda activate gadgetron), you should be able to check that everything is working with gadgetron --info

Using Docker containers

Docker images are built automatically from the Gadgetron project. The latest runtime images are:

ghcr.io/gadgetron/gadgetron/gadgetron_ubuntu_rt_cuda:latest: The latest runtime image with CUDA support.
ghcr.io/gadgetron/gadgetron/gadgetron_ubuntu_rt_nocuda:latest: The latest runtime image without CUDA support.

To run the Gadgetron:

docker run --gpus=all -ti -p 9004:9002 ghcr.io/gadgetron/gadgetron/gadgetron_ubuntu_rt_cuda:latest

This will run the GPU enabled version of the Gadgetron and expose it on port 9004.

For details on how to build these images yourself, see out build instructions

Running the Gadgetron in Kubernetes

The Docker images can be deployed in a Kubernetes cluster. See this repository for details.