The applications of ChatGPT and other large language models in anesthesiology and critical care: a systematic review
Authors' conclusion
While LLMs are not yet equipped to fully assist physicians in anesthesiology and critical care, they have significant potential, and their capabilities are rapidly improving.
Study context
- Population
- Studies investigating large language models (LLMs) (GPT-4, GPT-3.5, Google Bard [now Gemini], LLaMA, others) in anesthesiology or critical care45 papers included of 480
Methodological detailsPRISMA, Searched PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science from inception until 1 August 2024
- PRISMA
- Searched PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science from inception until 1 August 2024
Pooled result
As reported in the abstract
- Applications in 4 segments: ICU, patient education, medical education, perioperative care
- LLMs, especially newer models, promising in predicting clinical scores, navigating simple clinical scenarios, managing preoperative anxiety
- Performance below clinician level in predicting outcomes, complex clinical scenarios (i.e., airway management), board examinations, generating patient-directed documents; newer models better than older ones
Original publication
The applications of ChatGPT and other large language models in anesthesiology and critical care: a systematic review. Can J Anaesth. Published 16 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →