DUE: Dynamic Uncertainty-Aware Explanation Supervision via 3D Imputation
This work addresses the problem of improving deep learning model interpretability for medical professionals in 3D imaging, but it appears incremental as it extends existing explanation supervision methods to a higher-dimensional domain.
The paper tackled the problem of applying explanation supervision to 3D medical images, which is under-explored due to challenges like spatial correlation changes and lack of annotations, and proposed the DUE framework to enhance model predictability and explainability in medical imaging diagnosis.
Explanation supervision aims to enhance deep learning models by integrating additional signals to guide the generation of model explanations, showcasing notable improvements in both the predictability and explainability of the model. However, the application of explanation supervision to higher-dimensional data, such as 3D medical images, remains an under-explored domain. Challenges associated with supervising visual explanations in the presence of an additional dimension include: 1) spatial correlation changed, 2) lack of direct 3D annotations, and 3) uncertainty varies across different parts of the explanation. To address these challenges, we propose a Dynamic Uncertainty-aware Explanation supervision (DUE) framework for 3D explanation supervision that ensures uncertainty-aware explanation guidance when dealing with sparsely annotated 3D data with diffusion-based 3D interpolation. Our proposed framework is validated through comprehensive experiments on diverse real-world medical imaging datasets. The results demonstrate the effectiveness of our framework in enhancing the predictability and explainability of deep learning models in the context of medical imaging diagnosis applications.