Evidential Feature Representation of Heterogeneous Information
By integrating confidence modeling, evidence combination, and knowledge incorporation, this direction provides robust and interpretable feature representations for handling heterogeneous and low-quality medical data, significantly improving the accuracy and safety of downstream tasks.
Confidence Modeling
Focuses on constructing reliable confidence estimates for heterogeneous and noisy medical data, enabling robust uncertainty-aware feature representations.
Evidence Combination
Focuses on constructing reliable confidence estimates for heterogeneous and noisy medical data, enabling robust uncertainty-aware feature representations.
Knowledge Integration
Focuses on integrating domain and clinical knowledge into evidential representations, enhancing model interpretability, generalization, and safety in real-world medical applications.
