Segmentation of Locally Varying Numbers of Outer Retinal Layers by a Model Selection Approach.
Jelena NovoselSuzanne YzerKoenraad A. VermeerLucas J. van VlietPublished in: IEEE Trans. Medical Imaging (2017)
Keyphrases
- model selection
- varying numbers
- motion segmentation
- cross validation
- bayesian information criterion
- parameter estimation
- hyperparameters
- regression model
- optical coherence tomography
- machine learning
- bayesian learning
- sample size
- level set
- oct images
- mixture model
- statistical learning
- statistical inference
- image segmentation
- variable selection
- model selection criteria
- marginal likelihood
- information criterion
- segmentation algorithm
- error estimation
- segmentation method
- selection criterion
- gaussian process
- parameter determination
- leave one out cross validation
- generalization bounds
- energy function
- unsupervised segmentation
- prior information
- generalization error
- automatic model selection
- graph cuts