An automatic detection approach of traumatic bleeding based on 3D CNN networks

Keyphrases

• 3D Convolutional Neural Network (3D CNN) 100%
• 3D Patches 33%
• Automatic Detection 100%
• Bleeding 66%
• Bleeding Detection 33%
• Boundary Problem 33%
• Classification Network 33%
• Classification Process 33%
• CNN Network 100%
• Contrast-enhanced CT Images 33%
• Ct Value 33%
• Deep CNN Network 33%
• Detection Method 100%
• DICOM Images 33%
• High Sensitivity 66%
• Low Contrast 33%
• Low-contrast Target 33%
• Medical Doctors 33%
• Medical Field 33%
• Number of Tests 33%
• Point Cloud Data 33%
• Preprocessing Techniques 33%
• Processing Image 33%
• Segmentation Classification 33%
• Segmentation Method 66%
• Segmentation Network 33%
• Traumatic Bleeding 100%
• Window Settings 33%

Computer Science

• 3-Dimensional Convolutional Neural Networks 100%
• Automatic Detection 100%
• Classification Process 33%
• Convolutional Neural Network 33%
• Data Type 33%
• False Positive 100%
• Medical Doctor 33%
• Network Segmentation 33%
• Point Cloud 33%
• Positive Number 33%
• Positive Prediction 33%
• Preprocessing 33%
• Processing Approach 33%
• Segmentation Process 66%

Engineering

• Classification Network 33%
• Classification Process 33%
• Convolutional Neural Network 100%
• Data Format 33%
• Medical Doctor 33%
• Medical Field 33%
• Point Cloud 33%
• Positive Number 33%
• Processed Image 33%