Correction of Motion Artifacts Using a Multiscale Fully Convolutional Neural Network.

K Sommer Axel Saalbach T Brosch C Hall Nathan M Cross Jalal B Andre

Published in: AJNR. American journal of neuroradiology (2020)

Retrospective correction of motion artifacts using a multiscale fully convolutional network is promising and may mitigate the substantial motion-related problems in the clinical MRI workflow.

Keyphrases

convolutional neural network
high speed
deep learning
mental health
magnetic resonance imaging
image quality
contrast enhanced
cross sectional
computed tomography
magnetic resonance
machine learning
diffusion weighted imaging
neural network
network analysis