Pre- and Post-processing Strategies for Generic Slice-wise Segmentation of Tomographic 3D Datasets Utilizing U-Net Deep Learning Models Trained for Specific Diagnostic Domains.
Gerald A. ZwettlerWerner BackfriederDavid R. Holmes IIIPublished in: VISIGRAPP (5: VISAPP) (2020)
Keyphrases
- learning models
- pre and post processing
- segmentation algorithm
- image segmentation
- machine learning
- learning tasks
- conditional random fields
- semi supervised learning
- pairwise
- segmentation method
- learning problems
- level set
- loss function
- machine learning algorithms
- learning algorithm
- training set
- classification models
- data mining
- hidden markov models
- image reconstruction
- transfer learning
- object recognition
- bayesian networks
- training data