Multi-component deformable models coupled with 2D-3D U-Net for automated probabilistic segmentation of cardiac walls and blood.
Dong YangQiaoying HuangLeon AxelDimitris N. MetaxasPublished in: ISBI (2018)
Keyphrases
- deformable models
- multi component
- shape prior
- medical images
- image segmentation
- volumetric images
- cardiac mr
- image analysis
- statistical shape
- medical image analysis
- object boundaries
- deformable contour
- carotid artery
- atlas construction
- segmentation algorithm
- shape modeling
- snake model
- external force
- medical imaging
- object segmentation
- wall motion
- left ventricle
- level set
- segmentation method
- probabilistic model
- imaging modalities
- anatomical structures
- fully automated
- mr images
- clinical applications
- three dimensional
- automated segmentation
- displacement field
- patient specific
- energy functional
- magnetic resonance imaging
- magnetic resonance
- region growing
- ct images
- fully automatic
- non stationary
- shape variations
- active shape model
- magnetic resonance images
- auxiliary variables
- image registration