Perioperative medicine & outcomes
Real‐world deployment and evaluation of PEri ‐operative AI CHatbot ( PEACH ): a large language model chatbot for peri‐operative medicine
Authors' conclusion
PEACH is an accurate, adaptable tool that enhances consistency and efficiency in peri-operative decision-making.
Study context
- Population
- Real-world peri-operative clinical queries
- Participants
- 240 real-world clinical iterations
- Outcome
- Accuracy vs institutional guidelines and expert consensus
Methodological detailssilent deployment with real-world data
- Silent deployment with real-world data
Main finding
As reported in the abstract
Accuracy
First-generation accuracy (78/80)
97.5%
Overall accuracy across three iterations (232/240)
96.7%
Accuracy of updated PEACH (235/240) vs null hypothesis of 95% accuracy, p = 0.018
97.9%
Further findings
Usability
- Clinicians noted PEACH expedited decisions
- 95%
Inter-rater reliability (κ) between three iterations
- PEACH κ
- 0.772 and 0.893
- Experienced peri-operative physicians κ
- 0.610 and 0.784
Safety
- Hallucinations
- 1/240
- Deviations
- 2/240
- Minimal
Original publication
Real‐world deployment and evaluation of PEri ‐operative AI CHatbot ( PEACH ): a large language model chatbot for peri‐operative medicine. Anaesthesia. Published 19 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →