Exploring Entropy-based Active Learning for Fair Brain Segmentation

Keyphrases

• 3D U-Net 20%
• Active Learning 100%
• Active Learning Framework 20%
• Active Learning Methods 20%
• Active Learning Strategies 20%
• Bias Setting 20%
• Brain MRI 20%
• Brain Segmentation 100%
• Classification Task 20%
• Current Group 20%
• Efficient Strategy 20%
• Entropy-based Approach 100%
• Epistemic Uncertainty 20%
• Fairness-aware 20%
• Group-specific 20%
• Label Set 40%
• Learning Cycle 20%
• Left Caudate 40%
• Medical Image Segmentation 60%
• On-state Current 20%
• Performance Disparities 40%
• Performance Estimate 20%
• Performance Fairness 20%
• Performance Metrics 20%
• Region of Interest 20%
• Resource-constrained Environments 20%
• Scaling Entropy 20%
• Selection Strategy 20%
• Sensitive Attribute 20%
• Specific Performance 20%
• Standard Entropy 20%
• T1-weighted 20%
• Uncertainty Sampling 20%
• Weighted Entropy 20%

Computer Science

• Active Learning 100%
• Classification Task 12%
• Constrained Environment 12%
• Entropy 100%
• Image Segmentation 37%
• Learning Framework 12%
• Performance Metric 12%
• Selection Strategy 12%
• Sensitive Attribute 12%
• U-Net 12%

Social Sciences

• Entropy 100%
• Image Segmentation 75%
• Learning Method 25%
• Learning Strategy 25%
• Magnetic Resonance Imaging 25%
• U-Net 25%
• Variance 25%