Few-shot brain segmentation from weakly labeled data with deep heteroscedastic multi-task networks.
Richard McKinleyMichael RebsamenRaphael MeierMauricio ReyesChristian RummelRoland WiestPublished in: CoRR (2019)
Keyphrases
- multi task
- brain segmentation
- multi task learning
- learning tasks
- multi class
- learning problems
- feature selection
- sparse learning
- brain images
- semi supervised learning
- mr images
- anatomical structures
- weakly labeled data
- information gain
- machine learning
- transfer learning
- image data
- computer vision
- magnetic resonance imaging
- multi label
- maximum likelihood
- image processing