Multi-task learning for joint weakly-supervised segmentation and aortic arch anomaly classification in fetal cardiac MRI.
Paula Ramirez GillilandAlena UusMilou P. M. van PoppelIrina GrigorescuJohannes K. SteinwegDavid F. A. LloydKuberan PushparajahAndrew P. KingMaria DeprezPublished in: CoRR (2023)
Keyphrases
- cardiac mri
- multi task learning
- weakly supervised
- multi task
- feature extraction
- learning tasks
- image segmentation
- topic models
- machine learning
- decision trees
- left ventricle
- learning problems
- segmentation method
- image classification
- feature selection
- semi supervised
- training set
- generative model
- level set
- higher order
- multi class
- shape prior
- classification models
- support vector machine
- feature space
- support vector
- left ventricular
- multiscale