Weakly supervised 3D classification of chest CT using aggregated multi-resolution deep segmentation features.
Anindo SahaFakrul Islam TusharKhrystyna FarynaVincent M. D'AnniballeRui HouMaciej A. MazurowskiGeoffrey D. RubinJoseph Y. LoPublished in: Medical Imaging: Computer-Aided Diagnosis (2020)
Keyphrases
- weakly supervised
- fully supervised
- multiresolution
- classification accuracy
- feature vectors
- feature set
- feature extraction
- feature space
- chest ct
- segmentation algorithm
- image features
- superpixels
- co occurrence
- image segmentation
- level set
- segmentation method
- image analysis
- decision trees
- medical images
- image classification
- support vector
- feature selection
- lung nodules
- relation extraction
- feature subset
- active learning
- named entities
- image regions
- shape prior
- text classification
- object class
- training data
- multiscale
- training set
- computer vision