Meeting Banner
Abstract #3780

Correction of RF Spike Noise in MR Images Using Robust Principal Component Analysis

Adrienne E. Campbell-Washburn1, 2, David Atkinson3, Oliver Josephs4, Mark F. Lythgoe5, Roger J. Ordidge6, David L. Thomas7

1Centre for Advnaced Biomedical Imaging, University College London, London, United Kingdom; 2Department of Medical Physics and Bioengineering, University College London, London, United Kingdom; 3Centre for Medical Imaging and Centre for Medical Image Computing, University College London, London, United Kingdom; 4University College London and Birkbeck College, London, United Kingdom; 5Centre for Advanced Biomedical Imaging, Division of Medicine and Institute of Child Health, University College London, London, United Kingdom; 6Centre for Neuroscience, University of Melbourne, Melbourne, Victoria, Australia; 7Department of Brain Repair and Rehabilitation, UCL Institute of Neurology, University College London, London, United Kingdom

RF spike noise, caused by hardware problems, can lead to striping artefacts in MR images. These artefacts affect the image quality and quantitative information from the MRI data, and often must be removed in post-processing. This abstract presents an algorithm for semi-automated detection and correction of RF spike noise based on Robust Principal Component Analysis (RPCA). RPCA is used to decompose the measured k-space into low-rank (artefact-free) and sparse (RF spike) matrix components, including an automatic correction for the misidentification of the central k-space cluster. This algorithm is demonstrated to efficiently and effectively recover artefact-free data and regain quantitative information.

Abstract PDF Presentation Video

Keywords

accept added adding addition adjust advanced affect aims application applied applying arbitrarily audience automated biomedical bottom brain broader central choice cine classified clear clearly cluster coils college combination complex component components computing confirmation connections copyright correct corrected correction corrupted corruption decay decision decompose default detect detecting detection deviation differently distribution division effectively entries equal estimation example experimental frame frames free full hardware health heart ideally imperfections includes inexact institute intensity journal kidney kingdom mapping matrices matrix measured medical medicine mimic mouse multiplication multiplier must myocardial neurology noise norm nuclear object observation oliver optimal optimization original penalty perception perhaps positive post potential practice previous previously principal procedure processing quality quantitative random rank readout recovered recovery refilled refilling remain removed repair represents residual resonator returned robust scaled selection separating series sets shielding simulated since sometimes space sparse sparsity spike spiked spikes stripes striping synthesized tans target technical technologies transform type validate variation volume