A Novel Convolutional Neural Network Based on Adaptive Multi-Scale Aggregation and Boundary-Aware for Lateral Ventricle Segmentation on MR images.
Fei YeZhiqiang WangSheng ZhuXuanya LiKai HuPublished in: ICASSP (2022)
Keyphrases
- mr images
- segmentation result
- multiscale
- medical images
- accurate segmentation
- magnetic resonance
- brain mr images
- partial volume
- mri data
- deformable models
- brain tumors
- mr brain
- intensity inhomogeneity
- magnetic resonance images
- mr imaging
- brain structures
- manual segmentation
- mr brain images
- cardiac magnetic resonance
- bias field
- intensity distribution
- region of interest
- dice similarity coefficient
- image segmentation
- phantom images
- brain segmentation
- inhomogeneity correction
- image intensity
- object boundaries
- brain tissue
- cerebrospinal fluid
- contrast enhanced
- shape prior
- nonrigid registration
- quantitative measurements
- image data
- inter patient
- tissue segmentation
- level set
- edge detection
- region growing
- brain mri
- lesion segmentation
- segmentation algorithm
- automated segmentation
- mri images
- anatomical structures
- medical imaging
- gray matter
- medical image analysis
- image analysis
- fully automatic
- model based segmentation
- magnetic resonance imaging
- low contrast