Automatic deep learning-based segmentation of neonatal cerebral ventricles from 3D ultrasound images.
Zachary SzentimreySandrine de RibaupierreAaron FensterEranga UkwattaPublished in: Medical Imaging: Biomedical Applications in Molecular, Structural, and Functional Imaging (2021)
Keyphrases
- deep learning
- fully automatic
- segmentation algorithm
- segmentation method
- intensity variations
- test images
- brain mri
- input image
- image regions
- unsupervised learning
- image segmentation
- image features
- ground truth
- image classification
- bounding box
- region of interest
- image processing
- feature extraction
- mental models
- multiple images
- object recognition
- image retrieval
- medical images
- reinforcement learning
- shape prior
- human brain
- energy function
- superpixels
- weakly supervised
- multiscale
- graph cuts
- maximum likelihood