Comparison of a Hybrid Mixture Model and a CNN for the Segmentation of Myocardial Pathologies in Delayed Enhancement MRI.
Markus HüllebrandMatthias IvantsitsHannu ZhangPeter KohlmannJan-Martin KuhnigkTitus KühneStefan O. SchönbergAnja HennemuthPublished in: M&Ms and EMIDEC/STACOM@MICCAI (2020)
Keyphrases
- mixture model
- phase contrast
- medical images
- gaussian mixture model
- accurate segmentation
- em algorithm
- density estimation
- generative model
- bayesian information criterion
- mr images
- mri images
- probabilistic model
- maximum likelihood
- expectation maximization
- microscopy images
- magnetic resonance
- magnetic resonance imaging
- low contrast
- unsupervised learning
- model selection
- language model
- segmentation algorithm
- level set
- mixture modeling
- mri data
- image segmentation
- brain tumors
- carotid artery
- brain structures
- generalized em algorithm
- background subtraction
- medical imaging
- segmentation method
- magnetic resonance images
- left ventricular
- image processing
- tagged mri
- automatic segmentation
- probability density function
- short axis
- computer vision
- anatomical structures
- training data
- edge detection
- fully automatic
- manual segmentation
- left ventricle
- gaussian mixture