Impact of network parameters on a U-Net based system for rectal cancer segmentation on MR images.
Jovana PanicValentina GianniniArianna DefeudisDaniele ReggeGabriella BalestraSamanta RosatiPublished in: MeMeA (2022)
Keyphrases
- mr images
- prostate cancer
- network parameters
- accurate segmentation
- medical images
- brain mr images
- partial volume
- magnetic resonance
- brain tumors
- magnetic resonance images
- intensity inhomogeneity
- manual segmentation
- bias field
- brain structures
- dice similarity coefficient
- mr brain
- cardiac magnetic resonance
- prostate segmentation
- brain mri
- intensity distribution
- mr imaging
- image guided
- deformable registration
- inhomogeneity correction
- dce mri
- automated segmentation
- image data
- mri data
- conditional probabilities
- phantom images
- anatomical structures
- nonrigid registration
- tissue segmentation
- radiation therapy
- brain tissue
- inter patient
- medical image analysis
- contrast enhanced
- network size
- segmentation method
- shape prior
- medical imaging
- level set
- imaging modalities
- fully automatic
- deformable models
- image segmentation
- computer aided
- ct images
- model based segmentation
- clinical data
- treatment planning
- brain images