MetricUNet: Synergistic image- and voxel-level learning for precise prostate segmentation via online sampling

Computer Science

• Conventional Learning 33%
• Convolutional Network 66%
• Feature Space 33%
• Image Segmentation 33%
• Learned Feature 33%
• Multitask Learning 33%
• Network Architecture 33%
• Subnetwork 100%
• Training Phase 33%

Keyphrases

• Branch Architecture 33%
• Dice Loss 66%
• Dual-branch 33%
• Metric Loss 33%
• Multi-task Network 33%
• Online Metric Learning 33%
• U-Net Architecture 33%