Segmentation of retinal vessels in fundus images based on U-Net with self-calibrated convolutions and spatial attention modules.
Yibiao RongYu XiongChong LiYing ChenPeiwei WeiChuliang WeiZhun FanPublished in: Medical Biol. Eng. Comput. (2023)
Keyphrases
- fundus images
- blood vessels
- optic disc
- retinal blood vessels
- retinal images
- vascular tree
- optic disk
- optical coherence tomography
- retinal fundus images
- vessel segmentation
- vessel detection
- low contrast
- retinal vessel
- diabetic retinopathy
- cross sectional
- image segmentation
- accurate segmentation
- magnetic resonance angiography
- automatic detection
- region of interest
- cross sections
- segmentation accuracy
- level set
- multiscale