3-D Fully Convolutional Networks for Multimodal Isointense Infant Brain Image Segmentation

Keyphrases

• Brain Image Segmentation 100%
• Infant Brain Images 100%
• Fully Convolutional Network 100%
• Isointense Phase 80%
• Network Architecture 40%
• Tissue Segmentation 40%
• Segmentation Performance 40%
• Feature Map 40%
• Accurate Segmentation 20%
• Magnetic Resonance Imaging 20%
• High-resolution 20%
• Brain Magnetic Resonance Imaging 20%
• Patch-based 20%
• Brain MRI 20%
• State-of-the-art Techniques 20%
• Network Model 20%
• Gray Matter 20%
• White Matter 20%
• Region of Interest 20%
• Early Brain Development 20%
• Single Modality 20%
• Tissue Contrast 20%
• Coarse Features 20%
• Segmentation Accuracy 20%
• Feature Fusion 20%
• 2D to 3D 20%
• Myelination 20%
• Large Margin 20%
• Sparse Labels 20%
• Multimodal Information 20%
• Hierarchical Feature Aggregation 20%
• Fast Segmentation 20%
• Semantic Feature Map 20%
• Batch Normalization 20%
• Dense Features 20%
• Fusion Module 20%
• Transformation Module 20%

Computer Science

• Image Segmentation 100%
• Fully Convolutional Network 100%
• Feature Map 60%
• Network Architecture 40%
• Segmentation Performance 40%
• Semantic Feature 20%
• Speed-up 20%
• And-States 20%
• Comparison Result 20%
• segmentation accuracy 20%
• Multimodal Information 20%
• Batch Normalization 20%

Neuroscience

• Magnetic Resonance Imaging 100%
• Gray Matter 50%
• Brain Development 50%
• Myelinogenesis 50%