Automatic lower limb bone segmentation in radiographs with different orientations and fields of view based on a contextual network

Keyphrases

• Radiograph 100%
• Bone Segmentation 100%
• Lower Limb Bones 100%
• Contextual Network 100%
• View-based 100%
• Bone Structure 60%
• Detection Rate 40%
• Dice Coefficient 40%
• X-ray Images 40%
• Neural Network 20%
• Proposed Architecture 20%
• Orthopedics 20%
• Automatically Identify 20%
• Neural Network Architecture 20%
• Manual Operation 20%
• Label Map 20%
• Receptive Field 20%
• Local Contextual Information 20%
• Structure Structure 20%
• Variable Configuration 20%
• Bone Identification 20%
• Pixel-wise Semantic Segmentation 20%
• Clinical Procedures 20%
• Patient Orientation 20%
• SegNet 20%
• Overlapping Structure 20%
• Global Contextual Information 20%

Computer Science

• Average Detection Rate 100%
• Image Segmentation 50%
• Neural Network 50%
• Neural Network Architecture 50%
• Receptive Field 50%
• Configuration Variable 50%
• Contextual Information 50%