Application-driven Validation of Posteriors in Inverse Problems
This work addresses a critical gap in validating posterior-based methods for inverse problems, particularly in medical imaging, though it is incremental as it adapts existing validation concepts from object detection.
The authors tackled the lack of application-driven validation for posterior-based methods in inverse problems, where multiple plausible solutions exist, by introducing a systematic framework that adapts object detection validation principles to enable mode-centric validation with interpretable metrics, demonstrating advantages in synthetic and medical vision use cases.
Current deep learning-based solutions for image analysis tasks are commonly incapable of handling problems to which multiple different plausible solutions exist. In response, posterior-based methods such as conditional Diffusion Models and Invertible Neural Networks have emerged; however, their translation is hampered by a lack of research on adequate validation. In other words, the way progress is measured often does not reflect the needs of the driving practical application. Closing this gap in the literature, we present the first systematic framework for the application-driven validation of posterior-based methods in inverse problems. As a methodological novelty, it adopts key principles from the field of object detection validation, which has a long history of addressing the question of how to locate and match multiple object instances in an image. Treating modes as instances enables us to perform mode-centric validation, using well-interpretable metrics from the application perspective. We demonstrate the value of our framework through instantiations for a synthetic toy example and two medical vision use cases: pose estimation in surgery and imaging-based quantification of functional tissue parameters for diagnostics. Our framework offers key advantages over common approaches to posterior validation in all three examples and could thus revolutionize performance assessment in inverse problems.