A shape-supervised feature fusion U-Net for tubular structure segmentation.
Jinghua YueShuo JinSiyuan WangJianping ZengSiqiao ShanBo LiuNan JiangFugen ZhouPublished in: Comput. Electr. Eng. (2024)
Keyphrases
- tubular structures
- feature fusion
- minimal path
- blood vessels
- shape prior
- feature extraction
- multiple features
- semi supervised
- supervised learning
- active contour model
- learning algorithm
- carotid artery
- feature selection
- energy function
- segmentation algorithm
- level set
- energy functional
- low level features
- multi sensor
- shape analysis
- shape model
- deformable models
- graph cuts
- computer vision