A Persistent Homology-Based Topological Loss Function for Multi-class CNN Segmentation of Cardiac MRI.
Nick ByrneJames R. CloughGiovanni MontanaAndrew P. KingPublished in: M&Ms and EMIDEC/STACOM@MICCAI (2020)
Keyphrases
- particle filter
- multi class
- loss function
- cardiac mri
- persistent homology
- pairwise
- topological features
- morse theory
- physiological parameters
- boosting algorithms
- multi class classification
- left ventricle
- computational geometry
- support vector
- object detection
- binary classification
- support vector machine
- left ventricular
- regularization term
- magnetic resonance images
- multi class boosting
- level set
- binary classifiers
- cost sensitive
- image segmentation
- feature selection
- fully automatic
- segmentation algorithm
- markov random field
- multiscale
- training data