FORGE

Fast, Optimized Reconstruction via Gridding Engine

FORGE is a C++ library and a set of command-line programs for iterative MRI reconstruction. It reads ISMRMRD raw data and writes NIfTI images, with SENSE, shot-by-shot phase correction, field-map correction and low-rank reconstruction, on the CPU, Apple Silicon GPUs (Metal) and NVIDIA GPUs (CUDA).

forgeSense reconstructing a synthetic radial scan while forgeview shows progress bars, convergence sparklines, the log and a live image preview
forgeSense with the forgeview terminal monitor

Tutorials

Lessons for newcomers. Reconstruct your first image, then call the library from C++.

How-to guides

Recipes for specific jobs: building, multi-shot and low-rank reconstruction, GPUs, MPI.

Reference

Every program and option, the input and output formats, and the C++ API.

Explanation

How the operators, field correction, solver and GPU backends work, and why.

Quick start on a Mac

With the dependencies installed:

git clone https://github.com/mechneurolab/forge.git && cd forge
cmake -B build -DMPISupport=OFF
cmake --build build -j$(sysctl -n hw.logicalcpu)
cmake --install build --prefix $HOME/forge
export PATH="$HOME/forge/bin/metal:$HOME/forge/bin:$PATH"

GenSyntheticISMRMRD -o sense.h5
mkdir -p recon
forgeSense -i sense.h5 -o recon/ -F NUFFT -t 1 -n 10