On the Uncertain Single-View Depths in Colonoscopies
This work addresses the challenge of depth estimation in colonoscopies for AI-assisted medical technologies, but it is incremental as it applies existing Bayesian methods to a new domain.
The paper tackles the problem of single-view depth estimation in colonoscopies by exploring Bayesian deep networks for the first time in this domain, resulting in a novel teacher-student approach that incorporates teacher uncertainty and an analysis of scalable methods across datasets.
Estimating depth information from endoscopic images is a prerequisite for a wide set of AI-assisted technologies, such as accurate localization and measurement of tumors, or identification of non-inspected areas. As the domain specificity of colonoscopies -- deformable low-texture environments with fluids, poor lighting conditions and abrupt sensor motions -- pose challenges to multi-view 3D reconstructions, single-view depth learning stands out as a promising line of research. Depth learning can be extended in a Bayesian setting, which enables continual learning, improves decision making and can be used to compute confidence intervals or quantify uncertainty for in-body measurements. In this paper, we explore for the first time Bayesian deep networks for single-view depth estimation in colonoscopies. Our specific contribution is two-fold: 1) an exhaustive analysis of scalable Bayesian networks for depth learning in different datasets, highlighting challenges and conclusions regarding synthetic-to-real domain changes and supervised vs. self-supervised methods; and 2) a novel teacher-student approach to deep depth learning that takes into account the teacher uncertainty.