DEEP LEARNING TECHNIQUES FOR SUPPRESSING ARTEFACTS IN MAGNETIC RESONANCE IMAGES

摘要

Techniques for removing artefacts, such as RF interference and/or noise, from magnetic resonance data. The techniques include: obtaining (302) input magnetic resonance data using at least one radio-frequency coil (526) of a magnetic resonance imaging system (500); and generating (306) a magnetic resonance image from the input magnetic resonance data at least in part by using a neural network model (130) to suppress (304, 308) at least one artefact in the input magnetic resonance data.