AL-DRL achieves over 97% gastric coverage in patient-derived simulations within 50 seconds and 87% mean coverage in ex-vivo tests with 53% faster procedures than expert manual control.
EndoSLAM dataset and an unsupervised monoc- ular visual odometry and depth estimation approach
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Anatomical Landmark-Guided Deep Reinforcement Learning for Autonomous Gastric Navigation
AL-DRL achieves over 97% gastric coverage in patient-derived simulations within 50 seconds and 87% mean coverage in ex-vivo tests with 53% faster procedures than expert manual control.