Note: This software package is outdated, and has been superseded by the software available from here. We still provide access to this older software for legacy purposes, but if you’re interested in using LORAKS, we encourage you to update to the most recent version.
This page provides a MATLAB software implementation and examples for the Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS) technique originally described in:
J. P. Haldar.
Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS) for Constrained MRI.
IEEE Transactions on Medical Imaging 33:668-681, 2014.
[link] [preprint]
In addition to the MATLAB implementation of a LORAKS-based algorithm, the download also includes multiple demonstrations that illustrate the power and flexibility of the LORAKS framework in the contexts of MRI reconstruction and multiband signal reconstruction. The software and demonstrations are comprehensively described (including example outputs) in the following technical report:
J. P. Haldar.
Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS): Implementation and Examples for Reproducible Research.
University of Southern California, Los Angeles, CA,
Technical Report USC-SIPI-414, April 2014.
[link]
One example of an output from this software is shown below, demonstrating the ability of LORAKS to reconstruct sparsely-sampled data that is acquired with a highly-structured sampling pattern:
The software can be downloaded using the form below: