Non-Rigid Respiratory Motion Estimation of Whole-Heart Coronary MR Images Using Unsupervised Deep Learning.
Haikun QiNiccolo FuinGastão CruzJiazhen PanThomas KüstnerAurélien BustinRené M. BotnarClaudia PrietoPublished in: IEEE Trans. Medical Imaging (2021)
Keyphrases
- deep learning
- mr images
- motion estimation
- unsupervised learning
- motion compensation
- magnetic resonance
- cardiac magnetic resonance
- weakly supervised
- beating heart
- medical images
- magnetic resonance images
- motion vectors
- manual segmentation
- pre operative
- motion field
- cardiac motion
- supervised learning
- nonrigid registration
- machine learning
- mri data
- video sequences
- x ray
- optical flow
- image sequences
- super resolution
- semi supervised
- image data
- mr imaging
- respiratory motion
- computed tomography
- mental models
- imaging modalities
- prostate cancer
- feature selection
- object recognition
- model selection
- ct images
- medical imaging
- left ventricle
- superpixels
- contrast enhanced
- three dimensional
- short axis
- computer vision