We are making available a software package for denoising multi-echo spin-echo (MESE) MRI data with an approach known as Marchenko-Pastur principal component analysis (MP-PCA).
The software can be used on its own or as a preprocessing add-on to the EMC-based quantitative T2 mapping package and the quantitative multi-component T2 myelin-water fraction mapping package.
(Separately, we are also sharing a software for MP-PCA denoising of diffusion weighted data.)
By applying the MP-PCA denoising algorithm to complex or magnitude images (such as DICOM format images), the software cleans up noisy MESE MRI data prior to quantitative mapping of T2 or myelin-water fraction. The denoising leads to leading to more stable parametric maps and more reliable detection of subtle tissue changes, especially in regimes characterized by low signal-to-noise ratio (SNR).
The package is implemented in MATLAB and operates on five-dimensional MESE datasets (the dimensions are coils, slices, readout, phase‑encode, and echo-times). The software is operated via the MATLAB script denoise_multi_echo_data.m, with user-defined image data and window size. The outputs are denoised MESE images and maps of SNR gain.
Further information how to use the MP-PCA denoising for MESE MRI package can be found in the User_Manual_PCA_denoising.pdf file included in the resource download.
Related Publication
Mapping of magnetic resonance imaging’s transverse relaxation time at low signal-to-noise ratio using Bloch simulations and principal component analysis image denoising.
NMR Biomed. 2022 Dec;35(12):e4807. doi: 10.1002/nbm.4807
Please cite these publication if you are using MP-PCA denoising for MESE MRI in your research.
Get the Code
The software available on this page is provided free of charge and comes without any warranty. CAI²R and NYU Grossman School of Medicine do not take any liability for problems or damage of any kind resulting from the use of the files provided. Operation of the software is solely at the user’s own risk. The software developments provided are not medical products and must not be used for making diagnostic decisions.
The software is provided for non-commercial, academic use only. Usage or distribution of the software for commercial purpose is prohibited. All rights belong to the author (Noam Ben-Eliezer) and NYU Grossman School of Medicine. If you use the software for academic work, please give credit to the author in publications and cite the related publications. For information regarding commercial use of the package or the underlying algorithm, get in touch with Noam Ben-Eliezer.
Contact
Questions about this resource may be directed to Noam Ben-Eliezer, PhD at noambe@tauex.tau.ac.il.
Related Resources
Quantitative mapping of myelin content using multi-T2-component fitting of multi-echo spin-echo MRI data.
Software for quantification of T2 relaxation time and proton density from multi-echo spin-echo MRI data.
State-of-the-art denoising for diffusion and functional imaging applications.