A Unified Metric for Comprehensive and Consistent Multi-Organ Segmentation Evaluation in Clinical Settings

Our research aims to improve the comprehensiveness and consistency of model evaluation in clinical scenarios involving multi-index and multi-organ medical image segmentation. Unlike previous approaches, our proposed method aligns various accuracy metrics and confidence estimates across different multi-organ segmentation outcomes and integrates them into a unified metric. This approach reduces the complexity and uncertainty of the clinical evaluation model, providing a novel and concise standard for clinical assessment.