English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Gastroendoscopy has been a clinical standard for diagnosing and treatingconditions that affect a part of a patient's digestive system, such as thestomach. Despite the fact that gastroendoscopy has a lot of advantages forpatients, there exist some challenges for practitioners, such as the lack of 3Dperception, including the depth and the endoscope pose information. Suchchallenges make navigating the endoscope and localizing any found lesion in adigestive tract difficult. To tackle these problems, deep learning-basedapproaches have been proposed to provide monocular gastroendoscopy withadditional yet important depth and pose information. In this paper, we proposea novel supervised approach to train depth and pose estimation networks usingconsecutive endoscopy images to assist the endoscope navigation in the stomach.We firstly generate real depth and pose training data using our previouslyproposed whole stomach 3D reconstruction pipeline to avoid poor generalizationability between computer-generated (CG) models and real data for the stomach.In addition, we propose a novel generalized photometric loss function to avoidthe complicated process of finding proper weights for balancing the depth andthe pose loss terms, which is required for existing direct depth and posesupervision approaches. We then experimentally show that our proposedgeneralized loss performs better than existing direct supervision losses.

Learning-Based Depth and Pose Estimation for Monocular Endoscope with Loss Generalization