Deep metric learning with mirror attention and fine triplet loss for fundus image retrieval in ophthalmology

Keyphrases

• Ophthalmology 100%
• Fundus Image 100%
• Image Retrieval 100%
• Deep Metric Learning 100%
• Triplet Loss 100%
• Positive Pairs 60%
• Image Descriptors 60%
• Microaneurysms 40%
• Lesion Characteristics 40%
• Optic Disc 20%
• Diabetic Retinopathy 20%
• Medical Image 20%
• Diabetic Retinopathy Screening 20%
• Performance Improvement 20%
• Learning Framework 20%
• Ophthalmologist 20%
• Spatial Dependence 20%
• Evidence-based 20%
• Left Eye 20%
• Right Eye 20%
• Fundus 20%
• Comprehensive Experiment 20%
• Retrieval Practice 20%
• Ranking Performance 20%
• Subtle Differences 20%
• Discriminative Features 20%
• Similar Cases 20%
• Discriminative Information 20%
• Space Images 20%
• Spatial Structure Characteristics 20%
• Medicos 20%
• Macular Area 20%
• Small Lesions 20%
• Horizontal View 20%
• Visual Manifestations 20%
• Relevant Degree 20%
• Attention Score 20%
• Ranking Quality 20%
• Visual Descriptors 20%

Computer Science

• Image Retrieval 100%
• Performance Improvement 33%
• Learning Framework 33%
• Spatial Structure 33%
• retrieval performance 33%
• Discriminative Feature 33%
• Attention Score 33%