Evaluating the Performance of Large Language Models in Addressing Preoperative Patient Questions: A Systematic Review and Analysis
- Andrew Kelly (Department of Anaesthesiology, Galway University Hospital, Galway)
- Leanne Gaule (Department of Anaesthesiology, Galway University Hospital, Galway, Ireland)
- David Kelly (School of Medicine, University College Dublin, Dublin, Ireland)
- Róisín Cassidy (Department of Anaesthesiology, Galway University Hospital, Galway, Ireland)
- Hiliary Regan (Department of Anaesthesiology, Galway University Hospital, Galway, Ireland)
Abstract
Large Language Models (LLMs), including chatGPT, google gemini and microsoft co-pilot are increasingly explored in the perioperative setting. Integrated use of LLMs in healthcare has been shown to improve efficiency, accuracy and patient management. This systematic review assesses the performance of LLMs at answering patients preoperative questions across a range of surgical specialities.Keywords: LLMs, AI, artificial intelligence, patient management
How to Cite: Kelly, A., Gaule, L., Kelly, D., Cassidy, R., Regan, H., (2026). Evaluating the Performance of Large Language Models in Addressing Preoperative Patient Questions: A Systematic Review and Analysis. In: Abstracts of EBPOM Dingle, Ireland Conference 2025: 16-18 September 2025, Dingle, Ireland. Advances in Perioperative Care: UCL Press; 2026. Pp 1-3. DOI: 10.14324/111.444/apoc.ebpomdingle25.a3
Rights: Author, [2026]
Downloads:
Download PDF
View
PDF