MAC: An Artifact Correction Framework for Brain MRI based on Deep Neural Networks

Abstract

<jats:title>Abstract</jats:title><jats:p>The correction of artifacts in Magnetic Resonance Imaging (MRI) is crucial due to physiological phenomena and technical issues affecting diagnostic quality. Reverting from corrupted to artifact-free images is a complex task. Deep Learning (DL) models have been employed to preserve data characteristics and to identify and correct those artifacts. We propose<jats:bold>MAC</jats:bold>, a novel DL-based solution to correct artifacts in multi-contrast brain MRI scans.<jats:bold>MAC</jats:bold>offers two models: the simulation and the correction models. The simulation model introduces perturbations similar to those occurring in an exam while preserving the original image as ground truth; this is required as publicly available datasets rarely have motion-corrupted images. It allows the addition of three types of artifacts with different degrees of severity. The DL-based correction model adds a fourth contrast to state-of-the-art solutions while improving the overall performance of the models.<jats:bold>MAC</jats:bold>achieved the highest results in the FLAIR contrast, with a Structural Similarity Index Measure (SSIM) of 0.9803 and a Normalized Mutual Information (NMI) of 0.8030. Moreover, the model reduced training time by 63% compared to its predecessor.<jats:bold>MAC</jats:bold>model can correct large volumes of images faster and adapt to different levels of artifact severity than current state-ofthe-art models, allowing for better diagnosis.</jats:p>