Multi-task learning of a deep K-nearest neighbour network for histopathological image classification and retrieval

Keyphrases

• Nearest Neighbor 100%
• Multi-task Learning 100%
• Model Interpretability 100%
• Image Retrieval 100%
• Histopathological Image Classification 100%
• Neighborhood Networks 100%
• K-nearest 100%
• Human-Interpretable 50%
• Sampling Strategy 50%
• Deep Neural Network 50%
• Object Detection 50%
• Model Performance 50%
• Deep Learning 50%
• Medical Applications 50%
• Classification Detection 50%
• Loss Function 50%
• Colorectal Cancer 50%
• Hashing 50%
• Training Database 50%
• Classification Network 50%
• Similar Images 50%
• Deep Learning Framework 50%
• Confidence Estimation 50%
• Medical Records 50%
• Expert Evaluation 50%
• Clinical Decision-making 50%
• Multi-task Deep Learning 50%
• Image Object 50%
• Triplet Loss Function 50%
• Interpretable Explanation 50%
• Confidence Score 50%
• Daily Clinical Practice 50%
• Model Failure 50%
• Lack of Transparency 50%
• Cancer Histology 50%
• Neighborhood Search 50%
• Image Recognition Classification 50%
• Recognition Detection 50%
• Strong Correlation 50%
• Triple Sampling 50%
• Histological Slides 50%

Computer Science

• Image Classification 100%
• Interpretability 100%
• Multitask Learning 100%
• Histopathological Image 100%
• Image Retrieval 100%
• Deep Learning Method 100%
• Learning Framework 50%
• Object Detection 50%
• Deep Neural Network 50%
• Performance Model 50%
• Hashing 50%
• Model Failure 50%
• Confidence Score 50%
• Decision-Making 50%
• Medical Applications 50%