ScholarDeck

Randomized trialNPJ digital medicine · 2026

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

Hai-Long Bing, Yan Wang, Shi-Ying Li … Qin-Jun Chu

AnesthesiologyRandomized trial

Automated anesthesia system matched clinician-managed anesthesia on hypoxemia during gastrointestinal endoscopy

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

Hai-Long Bing … Qin-Jun Chu

NPJ digital medicine · 2026

Background

The increasing demand for gastrointestinal endoscopic procedures, coupled with a global shortage of anesthesiologists, underscores the need for intelligent automation in anesthesia care.

Methods

A total of 509 participants were included in algorithm development, and 418 were enrolled for clinical validation.

n = 509 participants

Results

low blood oxygen was about equally likely with the automated system as with clinicians

OR 1.01 (95% CI 0.59 to 1.75)

null 1

0.59

1.75

CI crosses the null - not significant

More results

Reinforcement learning (reinforcement learning) offers a promising strategy for autonomous anesthesia control, yet prospective clinical validation remains limited.

More results

We developed an reinforcement learning-based automated anesthesia system for gastrointestinal endoscopy (anesthesia system for gastrointestinal endoscopy) for automated ciprofol delivery and conducted a prospective, multicenter, randomized controlled trial across four centers in China between January 8 and August 27, 2025.

“

Conclusion

Clinical registration: ClinicalTrials.gov on Feb. 26, 2025 (NCT06857344).

Post

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

Automated anesthesia system matched clinician-managed anesthesia on hypoxemia during gastrointestinal endoscopy

Related papers