Combining Shape Priors with Conditional Adversarial Networks for Improved Scapula Segmentation in Mr Images.
Arnaud BoutillonBhushan BorotikarValérie BurdinPierre-Henri ConzePublished in: ISBI (2020)
Keyphrases
- shape prior
- mr images
- medical images
- deformable models
- segmentation result
- magnetic resonance
- level set
- accurate segmentation
- graph cuts
- object segmentation
- active contours
- image segmentation
- object boundaries
- intensity distribution
- magnetic resonance images
- shape model
- segmentation method
- segmentation algorithm
- variational segmentation
- intensity inhomogeneity
- manual segmentation
- level set based segmentation
- curve evolution
- mri data
- mr imaging
- brain tumors
- active shape model
- image data
- brain mr images
- brain structures
- active contour model
- shape representation
- intensity information
- bayesian framework
- prior information
- bias field
- energy function
- prior knowledge
- shape variations
- global optimization
- user interaction
- appearance cues
- statistical models
- partial volume
- medical imaging
- level set method
- energy functional
- image intensity
- automatic segmentation
- cardiac magnetic resonance