FSS-2019-nCov: A deep learning architecture for semi-supervised few-shot segmentation of COVID-19 infection

Keyphrases

• COVID-19 100%
• Deep Learning Architectures 100%
• COVID-19 Infection 100%
• Few-shot Segmentation 100%
• Deep Learning 50%
• Encoder-decoder 50%
• COVID-19 CT 50%
• Feature Encoder 50%
• Context Enrichment 50%
• Recombination 25%
• Efficient Collaboration 25%
• Image Segmentation 25%
• Computed Tomography 25%
• Feature Extraction 25%
• Lung Infection 25%
• Learning Paradigm 25%
• Knowledge Exchange 25%
• Knowledge Representation 25%
• Labeled Data 25%
• Computational Complexity 25%
• Disease Diagnosis 25%
• High-level Information 25%
• Encoder 25%
• Lung Disease 25%
• Channel Information 25%
• CT Images 25%
• Coronavirus COVID-19 25%
• Efficient Segmentation 25%
• Recalibration 25%
• Accurate Segmentation 25%
• Segmentation Approach 25%
• Disease Quantification 25%
• Spatial Pyramid Pooling 25%
• Atrous Convolution 25%
• Convolution Block 25%
• CT Slices 25%
• Conditioner 25%
• Dual-path 25%
• 2019-nCoV Infection 25%
• Lung CT 25%
• Relevant Knowledge 25%
• Main Module 25%
• Few-shot Learning 25%
• Squeeze-and-excitation Network (SENet) 25%
• Multi-class Labeling 25%
• Multi-scale Pyramid 25%
• COVID-19 Pneumonia 25%
• Encoder Module 25%

Computer Science

• Deep Learning Method 100%
• Image Segmentation 50%
• Feature Extraction 50%
• Computational Complexity 50%
• Information Channel 50%
• Deep Learning Technique 50%
• Knowledge Exchange 50%
• Segmentation Approach 50%
• Few-Shot Learning 50%
• Multiscale Pyramid 50%