Skull segmentation from MR scans using a higher-order shape model based on convolutional restricted Boltzmann machines.
Oula PuontiKoen Van LeemputJesper D. NielsenChristian BauerHartwig Roman SiebnerKristoffer Hougaard MadsenAxel ThielscherPublished in: Medical Imaging: Image Processing (2018)
Keyphrases
- restricted boltzmann machine
- higher order
- conditional random fields
- deep learning
- shape prior
- medical images
- deep belief networks
- face shape
- deformable models
- brain images
- semi supervised
- segmentation method
- probabilistic graphical models
- learning algorithm
- markov random field
- shape model
- contrastive divergence
- level set
- brain tumors
- energy function
- multiscale
- unsupervised learning
- segmentation algorithm
- magnetic resonance
- image segmentation
- loopy belief propagation
- prior information
- ct images
- natural images
- pairwise
- cerebrospinal fluid
- object segmentation
- active contours
- sparse coding
- graphical models
- probabilistic model
- optical flow
- prior knowledge
- mr images
- machine learning