Combining prospective and retrospective motion correction based on a model for fast continuous motion.

Patrick Hucker Michael Dacko Maxim Zaitsev

Published in: Magnetic resonance in medicine (2021)

The combined correction results in better images than pure PMC in the presence of fast motion. The predictions of artifacts are very accurate, allowing for comparing sequences or protocols in simulations. The observed artifacts due to Nyquist violations are expected to be corrected by utilizing parallel imaging.

Keyphrases

high resolution
high speed
deep learning
molecular dynamics
cross sectional
optical coherence tomography
convolutional neural network
machine learning
cone beam
fluorescence imaging
genetic diversity