Reinterpreting the Transformation Posterior in Probabilistic Image Registration
This work addresses a foundational issue in medical imaging and computer vision by rethinking uncertainty estimation in image registration, though it appears incremental as it builds on existing probabilistic methods.
The paper challenges the conventional interpretation of transformation posteriors in probabilistic image registration, arguing that using the mode as the most likely transformation and summary statistics for uncertainty is unjustified. It proposes ensemble fields to encode variability, demonstrating their potential as a complement to registered images and a foundation for advanced uncertainty characterization.
Probabilistic image registration methods estimate the posterior distribution of transformation. The conventional way of interpreting the transformation posterior is to use the mode as the most likely transformation and assign its corresponding intensity to the registered voxel. Meanwhile, summary statistics of the posterior are employed to evaluate the registration uncertainty, that is the trustworthiness of the registered image. Despite the wide acceptance, this convention has never been justified. In this paper, based on illustrative examples, we question the correctness and usefulness of conventional methods. In order to faithfully translate the transformation posterior, we propose to encode the variability of values into a novel data type called ensemble fields. Ensemble fields can serve as a complement to the registered image and a foundation for developing advanced methods to characterize the uncertainty in registration-based tasks. We demonstrate the potential of ensemble fields by pilot examples