3D deeply supervised network for automated segmentation of volumetric medical images

Keyphrases

• Automatic Segmentation 100%
• Volumetric Medical Images 100%
• Deep Supervision Mechanism 100%
• Deeply Supervised Network 100%
• Convergence Rate 50%
• Upper Layer 50%
• Optimization Problem 50%
• Adverse Effects 50%
• Post-processing Techniques 50%
• Fast Speed 50%
• Lower Layer 50%
• Training Samples 50%
• 3D Network 50%
• Gradient Change 50%
• Training Procedure 50%
• Overfitting 50%
• Discrimination Power 50%
• Limited Training Data 50%
• Segmentation Problem 50%
• Redundant Computation 50%
• Vessel Segmentation 50%
• Grand Challenges 50%
• Gradient Vanishing 50%
• Medical Image Segmentation 50%
• Convolutional Neural Network 50%
• Deep Model 50%
• Explode 50%
• Volumetric Images 50%
• Image Optimization 50%
• Deep Convolutional Neural Network (deep CNN) 50%
• 3D Fully Convolutional Networks 50%
• MICCAI 50%
• Fully Connected Conditional Random Field 50%
• Whole-heart 50%
• Liver Segmentation 50%
• Great Vessels 50%
• 3D MR Images 50%
• 3D Medical Images 50%
• Unstable Gradient 50%
• Conditional Random Field Model 50%
• 3D CT Scan 50%
• Volumetric Medical Image Segmentation 50%

Computer Science

• Image Segmentation 100%
• Optimization Problem 50%
• Convolutional Neural Network 50%
• Vessel Segmentation 50%
• Postprocessing Step 50%
• Discrimination Capability 50%
• Conditional Random Field 50%
• Deep Convolutional Neural Networks 50%
• Segmentation Task 50%
• Liver Segmentation 50%
• Training Data 50%
• Objective Function 50%
• Grand Challenge 50%
• Convergence Speed 50%
• Training Sample 50%
• Random Field Model 50%
• Fully Convolutional Network 50%

Engineering

• Medical Image 100%
• Medical Image Segmentation 100%
• Convolutional Neural Network 100%
• Optimisation Problem 50%
• Processing Step 50%
• Convergence Speed 50%
• Random Field Model 50%
• Objective Function 50%