A Methodology to Train a Convolutional Neural Network-Based Low-Dose CT Denoiser With an Accurate Image Domain Noise Insertion Technique.
Byeongjoon KimSarah E. DivelNorbert J. PelcJongduk BaekPublished in: IEEE Access (2022)
Keyphrases
- low dose
- dual energy
- contrast enhancement
- computer tomography
- x ray
- signal dependent noise
- computed tomography
- ct images
- medical images
- imaging modalities
- ct scans
- image features
- image data
- lung nodules
- image analysis
- multiscale
- additive noise
- mri images
- region of interest
- image enhancement
- image segmentation
- three dimensional
- medical imaging
- input image
- low contrast
- edge detection
- clinical applications
- ct data
- prior information
- accurate registration
- image reconstruction
- gray level