Artificial intelligence in anesthesiology is evolving from narrow task-specific tools to large, generalizable foundation models capable of integrating multimodal perioperative data.
Artificial intelligence is increasingly applied in anesthesiology for perioperative decision-making, with future directions pointing toward multimodal foundation models and closed-loop control systems.
This narrative review examined how artificial intelligence is increasingly being applied in anesthesiology to support clinical decision-making across the perioperative period. It outlines current applications of artificial intelligence in preoperative risk assessment, intraoperative monitoring and automation, and postoperative complication prediction. We also examined the underlying artificial intelligence architectures that form the technical foundations of these tools, including machine learning, deep learning, and natural language processing. We propose that in the future, rather than narrow task-specific tools, artificial intelligence in anesthesiology should involve the development and clinical translation of large, generalizable foundation models capable of integrating multimodal perioperative data. In addition, developments in multimodal data integration, closed-loop control systems, and interpretable modeling may further refine these approaches. Further progress in artificial intelligence–driven anesthesiology may require multidisciplinary collaboration, prospective clinical validation, and careful integration into perioperative workflows to ensure safe and clinically meaningful adoption.
Zhang et al. (Mon,) conducted a review in Anesthesiology. Artificial intelligence was evaluated. Artificial intelligence in anesthesiology is evolving from narrow task-specific tools to large, generalizable foundation models capable of integrating multimodal perioperative data.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: