A Voxel-Based Fully Convolution Network and Continuous Max-Flow for Carotid Vessel-Wall-Volume Segmentation From 3D Ultrasound Images.
Ran ZhouFumin GuoM. Reza AzarpazhoohJohn David SpenceEranga UkwattaMingyue DingAaron FensterPublished in: IEEE Trans. Medical Imaging (2020)
Keyphrases
- segmentation algorithm
- segmentation method
- max flow
- ultrasound images
- image analysis
- accurate segmentation
- edge detection
- intravascular ultrasound
- input image
- image data
- fully automatic
- vessel wall
- carotid artery
- image segmentation
- speckle noise
- pixel level
- segmentation result
- microscopy images
- region growing
- image processing
- magnetic resonance angiography
- contrast agent
- image registration
- gray value
- level set
- contrast enhanced
- superpixels
- energy minimization
- markov random field
- energy function