opengate 10.0.0

Simulation for Medical Physics

This is GATE 10

GATE is an open-source Monte Carlo simulation platform designed for modeling and simulating medical imaging systems, radiotherapy, and radiation dosimetry. It extends Geant4 with features tailored for time-dependent simulations, imaging physics, and therapy modeling. Widely used in medical physics research, GATE supports advanced applications such as PET, SPECT, CT, internal and external beam radiotherapy. GATE fosters collaboration through its active open-source community, enabling continuous development and shared innovation.

GATE 10 introduces a Python-based interface, replacing the macro scripting of GATE 9.x, offering improved flexibility, readability, and integration with modern scientific workflows. Read our motivation.

• Documentation: see the User Guide.
• This current version uses Geant4 11.2.1.
• Compatible with Python 3.9, 3.10, 3.11, 3.12. (Not python 3.13 yet)
• Warning: on Windows, the multithreading and Qt visualization are not (yet) available.

How to install (short version)

First, create a python environment:

python -m venv opengate_env
source opengate_env/bin/activate
pip install --upgrade pip

Then install the package opengate. The associated package opengate_core is automatically downloaded. opengate_core installs Geant4 librairies.

pip install opengate

If you already installed the packages and want to upgrade to the latest version:

pip install --upgrade opengate

Once installed, you can run all tests:

opengate_tests

WARNING (1) The first time you run this command, the geant4 data and the test data will be downloaded. If the download fails (on some systems), try to add the following command before running opengate_tests:

export GIT_SSL_NO_VERIFY=1

All tests are in the folder here. The test data (binary files) are stored as a git submodule here: https://gitlab.in2p3.fr/opengamgate/gam_tests_data.

WARNING (2) Some tests (e.g. test034) needs gaga-phsp which needs pytorch that cannot really be automatically installed by the previous pip install (at least we don't know how to do). So, in order to run those tests, you will have to install both PyTorch and gaga-phsp first with:

pip install torch
pip install gaga-phsp

WARNING (3) With some linux systems (not all), you may encounter an error similar to “cannot allocate memory in static TLS block”. In that case, you must add a specific path to the linker as follows:

export LD_PRELOAD=<path to libG4processes>:<path to libG4geometry>:${LD_PRELOAD}

or

export GLIBC_TUNABLES=glibc.rtld.optional_static_tls=2000000

The test history can be visualized here: https://opengate.github.io/opengate_tests_results

How to install (long version, for developers)

See the developer guide

Included third party C++ libraries

• https://github.com/pybind/pybind11
• https://github.com/fmtlib/fmt
• https://github.com/p-ranav/indicators