Dissecting Model Failures in Abdominal Aortic Aneurysm Segmentation through Explainability-Driven Analysis

1. 1Compute dense attribution-based encoder focus map
2. 2Analyze XAI field to identify irrelevant structures
3. 3Adjust model training to focus on relevant targets
4. 4Evaluate model performance using explainability-driven metrics

Data scientists and AI engineers working on medical imaging projects can benefit from this research to improve model performance and identify failures

💡 Explainable AI (XAI) can help identify model failures by analyzing where the model focuses

🔍 Explainability-driven analysis improves abdominal aortic aneurysm segmentation models