AI Science Tracker
EN

Department of Anesthesiology and Perioperative Medicine, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China

1 discovery from 2026

Discoveries

Department of Anesthesiology and Perioperative Medicine, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China; Zhengzhou Engineering Research Center for Medical Data Elements, Zhengzhou, China; Department of Anesthesiology and Perioperative Medicine, Henan Provincial People’s Hospital, People’s Hospital of Zhengzhou University, Zhengzhou, China; Department of Anesthesiology, Anxi Hospital, Anxi, China; Department of Anesthesiology, Jinjiang Municipal Hospital (Shanghai Sixth People’s Hospital Fujian Hospital), Jinjiang, China; Department of Anesthesiology, Renmin Hospital of Hubei University of Medicine, Shiyan, China; Institute for Interdisciplinary Information Sciences, Tsinghua University, Beijing, Beijing, China; Faculty of Applied Sciences, Macao Polytechnic University, Macao, China; Intelligent Medical Research Center, Shenzhen United Scheme Technology Company Limited, Boston, USA; Biomedical Engineering Center, Shenzhen United Scheme Technology Company Limited, Shenzhen, Guangdong, China; Institute of Trauma and Metabolism, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China; Tianjian Laboratory of Advanced Biomedical Sciences, Academy of Medical Sciences, Zhengzhou University, Zhengzhou, China; Cell Research and Transformation Center, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China; Information Technology Center/ Big Data AI Center, Zhengzhou Central Hospital Affiliated to Zhengzhou University, Zhengzhou, China

Reinforcement learning based automated anesthesia system for gastrointestinal endoscopy with a multicenter randomized trial.

Source check: Source checked. Not independent scientific confirmation.

Finding: In a multicenter randomized trial of 418 adults undergoing endoscopy, AAS-GE and clinician-managed anesthesia had similar observed hypoxemia rates (14.42% vs 14.29%; OR 1.01, 95% CI 0.59–1.75; P=.968). Induction was shorter, while body movement was more frequent with AAS-GE. The published protocol powered a test of a difference based on expected rates of 20% versus 10%; it did not specify a noninferiority margin, so the trial does not establish that a prespecified noninferiority criterion was met.

Technology: Offline RL learned the dose-control policy; the model was selected and refined before the randomized evaluation

Publication: Peer-reviewed paper · Test context: prospective or independent check. Catalog categories, not an evidence grade.

AI contribution
Offline RL learned the dose-control policy; the model was selected and refined before the randomized evaluation.
Test and result
The primary report gives hypoxemia rates of 14.42% vs 14.29% (OR 1.01, 95% CI 0.59–1.75; P=.968), shorter induction, and more body movement with AAS-GE. The published protocol's sample-size calculation tests a difference assuming expected rates of 20% vs 10%; no noninferiority margin is specified.
Limitation / still unknown
A prespecified noninferiority margin and corresponding noninferiority analysis were not identified in the published protocol or primary report; the observed nonsignificant difference does not itself establish noninferiority.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://www.nature.com/articles/s41746-026-02657-8 · https://ia.intanes.org/protocol/25001.html

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 4/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.