Abdominal Aortic Aneurysm Segmentation Using Convolutional Neural Networks Trained with Images Generated with a Synthetic Shape Model.
Karen López-LinaresMaialen StephensInmaculada GarcíaIván MacíaMiguel Ángel González BallesterRaúl San José EstéparPublished in: MLMECH/CVII-STENT@MICCAI (2019)
Keyphrases
- shape model
- convolutional network
- prior shape knowledge
- convolutional neural networks
- shape prior
- segmentation algorithm
- shape variations
- active shape
- set of training images
- shape variability
- facial landmarks
- segmentation method
- deformable shapes
- test images
- image database
- image analysis
- level set segmentation
- energy function
- image data
- abdominal aortic aneurysm
- active shape model
- shape constraints
- ground truth
- three dimensional
- statistical shape model
- global shape
- input image
- shape parameters
- image registration
- geometric shapes
- object segmentation
- shape similarity
- low contrast
- partial occlusion
- level set
- image retrieval
- multiscale
- image segmentation
- point distribution model
- ground truth data
- computer vision
- active contours
- graph cuts