A hybrid approach for mammary gland segmentation on CT images by embedding visual explanations from a deep learning classifier into a Bayesian inference.
Xiangrong ZhouSeiya YamagishiTakeshi HaraHiroshi FujitaPublished in: Medical Imaging: Computer-Aided Diagnosis (2021)
Keyphrases
- bayesian inference
- ct images
- deep learning
- medical images
- medical imaging
- pet ct
- lung nodules
- prior information
- fracture detection
- computed tomography
- ct scans
- probabilistic model
- medical diagnosis
- computer aided detection
- lung parenchyma
- unsupervised learning
- region of interest
- image segmentation
- ground glass opacity
- automated segmentation
- image analysis
- machine learning
- particle filter
- level set
- pulmonary nodules
- prostate cancer
- image processing
- shape prior
- segmentation algorithm
- image registration
- mr images
- computer aided diagnosis
- x ray
- training set
- expectation maximization
- weakly supervised
- dimensionality reduction
- higher order
- feature space
- support vector
- training data
- feature extraction
- decision trees