Large language models in anaesthesiology: use of ChatGPT for American Society of Anesthesiologists physical status classification

Editor—Large language models (LLMs) are advanced artificial intelligence (AI) models. They include ChatGPT, 1 ChatGPT. Available from https://chat.openai.com/chat (accessed 13 July 2023). Google Scholar a language model that is optimised for natural language input based on the Generative Pre-trained Transformer (GPT) developed by OpenAI. We hypothesise that the strengths of GPT in text summarisation can be applied to the analysis of patient profiles, 2 Seibert K. Domhoff D. Bruch D. et al. Application scenarios for artificial intelligence in nursing care: rapid review. J Med Internet Res. 2021; 23e26522 Crossref PubMed Scopus (28) Google Scholar , 3 Beegle C. Hasani N. Maass-Moreno R. Saboury B. Siegel E. Artificial intelligence and positron emission tomography imaging workflow: technologists’ perspective. PET Clin. 2022; 17: 31-39 Abstract Full Text Full Text PDF PubMed Scopus (3) Google Scholar , 4 Lorkowski J. Grzegorowska O. Pokorski M. Artificial intelligence in the healthcare system: an overview. Adv Exp Med Biol. 2021; 1335: 1-10 Crossref PubMed Scopus (12) Google Scholar such as assigning American Society of Anesthesiologists physical status (ASA-PS), 5 ASA Physical Status Classification System. Available from https://www.asahq.org/standards-and-guidelines/asa-physical-status-classification-system (accessed 13 July 2023). Google Scholar which is a high-level summary of overall patient clinical state. Our study explored this novel use of GPT by scoring ASA-PS on a set of standardised patient scripts. GPT responded fluently to the scripts, had comparable ASA-PS scoring to humans in most cases, and had more consistent responses than anaesthesiologists.