Semi-Supervised Learning for Low-Dose CT Image Restoration with Hierarchical Deep Generative Adversarial Network (HD-GAN).
Kihwan ChoiMalinda VaniaSungwon KimPublished in: EMBC (2019)
Keyphrases
- image restoration
- semi supervised learning
- low dose
- unsupervised learning
- x ray
- semi supervised
- computed tomography
- unlabeled data
- labeled data
- ct scans
- total variation
- lung nodules
- edge preserving
- regularization parameter
- imaging modalities
- supervised learning
- machine learning
- markov random field
- computer tomography
- super resolution
- image processing
- contrast enhancement
- active learning
- training data
- medical images
- image reconstruction
- ct images
- gaussian noise
- medical imaging
- labeled and unlabeled data
- multiscale
- regularization framework
- generative model
- computational complexity