BART Workshop Materials

This repository contains information and demos for the Berkeley Advanced Reconstruction Toolbox (BART). This material will be presented at the 2019 ISMRM Annual Meeting, Montreal, Canada. Previously, the material was presented at the 2016 ISMRM Workshop on Data Sampling & Image Reconstruction.

NEW Run demos in the browser through MyBinder

Click the icon

Purpose

The purpose of this repository is to host and share demos and workshop materials for BART. From the website:

The Berkeley Advanced Reconstruction Toolbox (BART) is a free and open-source image-reconstruction framework for Computational Magnetic Resonance Imaging. It consists of a programming library and a toolbox of command-line programs. The library provides common operations on multi-dimensional arrays, Fourier and wavelet transforms, as well as generic implementations of iterative optimization algorithms. The command-line tools provide direct access to basic operations on multi-dimensional arrays as well as efficient implementations of many calibration and reconstruction algorithms for parallel imaging and compressed sensing.

Getting Started

The most up-to-date information can be found at the official BART website: http://mrirecon.github.io/bart.

The workshop material was tested with BART version 0.4.04

Download

The source code is available at https://github.com/mrirecon/bart/archive/v0.4.04.tar.gz. Untar and navigate to the bart directory:

wget https://github.com/mrirecon/bart/archive/v0.4.04.tar.gz
tar -xvvf v0.4.04.tar.gz && mv bart-0.4.04 bart
cd bart

Quick Installation

See the Quick-Install guide for quick installation instructions.

Demo dependencies

The demos use Jupyter notebooks with Python and Bash kernels. This is easiest to accomplish with Anaconda. After installing Anaconda, install the dependencies:

pip install h5py numpy matplotlib bash_kernel 

Finish installing the Bash kernel with the command,

python -m bash_kernel.install

Demos

The demos directory contains standalone demos that show different BART use cases. The demos are self-documented within each directory, and are summarized below.

1. Introduction to the BART command-line tools (intro)
2. Compute g-factor using Python and BART (gfactor)
3. Build a non-Cartesian SENSE reconstruction tool with the BART C API (sense-recon)

Additional demos:

1. Simulate phantom data and compare regularized reconstructions (pics-phantom)
2. Reconstruct an axial slice of dynamic contrast enhanced (DCE) data (pics-dce)
3. Build a GRASP reconstruction tool with bash scripting and BART command-line tools (grasp)
4. Use the Wave-CS reconstruction tool using the BART Matlab API (wave)