Improving the Segmentation of Anatomical Structures in Chest Radiographs Using U-Net with an ImageNet Pre-trained Encoder.
Maayan Frid-AdarAvi Ben-CohenRula AmerHayit GreenspanPublished in: RAMBO+BIA+TIA@MICCAI (2018)
Keyphrases
- anatomical structures
- medical images
- medical imaging
- pre trained
- chest radiographs
- x ray
- x ray images
- lung nodules
- ct images
- shape model
- deformable models
- image segmentation
- shape analysis
- segmentation method
- level set
- image processing
- shape prior
- segmentation algorithm
- training examples
- energy function
- image registration
- image analysis
- training data
- active learning
- prior knowledge
- computer aided diagnosis
- multiscale
- active contours
- decision trees