Advancing Perioperative Medicine With AI: Current Landscape and Research Priorities.
- 1Department of Medicine, Stanford University, 300 Pasteur Drive, Palo Alto, CA, 94304, United States, 1 6507234000.
- 2Department of Anesthesiology, Cleveland Clinic, Cleveland Clinic Lerner College of Medicine, Cleveland, OH, United States.
Abstract
Unlabelled
AI is increasingly being applied across the perioperative continuum. AI should be viewed as a technology that can reduce cognitive burden, collate fragmented data into clinically meaningful information, support decision-making, and extend precision education to clinicians and patients. This editorial accompanies the JMIR Perioperative Medicine webinar held in May 2026 and presents our perspectives-informed by our clinical experience-on where AI can assist perioperative clinicians and surgical patients, the evidence available today, what is still needed, and how clinicians can prepare to practice alongside AI tools. We consider opportunities in preoperative risk assessment and optimization; intraoperative intelligent decision support; emerging technologies, such as surgical digital twins; and postoperative recovery, monitoring, and prediction of clinical deterioration. Most AI applications remain early and require large-scale, prospective validation across diverse populations and resource settings. Realizing their potential also requires confronting substantial challenges, including limited and fragmented training data, the specialized models needed to interpret the large volumes of data captured in operative videos and during physiological monitoring, biases embedded in data and algorithms or biases inherent in human-AI interactions, and the determination of where human oversight is essential and where it is not feasible. When used responsibly, with human judgment remaining paramount, AI offers an unprecedented opportunity to advance perioperative care.
Related
- A modified Delphi study to determine Australian healthcare consumer priorities and consensus for the use of genomics and precision medicine approaches in pregnancy healthcare and research
- Advancing equitable pain care worldwide requires progress in low- and middle-income settings: priorities for practice, research, and policy
- Cardiometabolic medicine in China: Burden, challenges, and strategic priorities
- Clinical research landscape and evidence gaps in Traditional Chinese Medicine for recurrent urinary tract infections: a scoping review
- SFAR's position: Organization of perinatal care in anesthesia, critical care, and perioperative medicine in France
BibTeX
@article{rohatgi2026advancing,
title = {{Advancing Perioperative Medicine With AI: Current Landscape and Research Priorities.}},
author = {Nidhi Rohatgi and Piyush Mathur},
journal = {JMIR perioperative medicine},
year = {2026},
month = oct,
doi = {10.2196/107155},
url = {https://europepmc.org/article/MED/42961855},
note = {CC BY},
}
RIS
TY - JOUR
TI - Advancing Perioperative Medicine With AI: Current Landscape and Research Priorities.
AU - Nidhi Rohatgi
AU - Piyush Mathur
PY - 2026
DA - 2026/10/30
JO - JMIR perioperative medicine
DO - 10.2196/107155
UR - https://europepmc.org/article/MED/42961855
AB - Unlabelled AI is increasingly being applied across the perioperative continuum. AI should be viewed as a technology that can reduce cognitive burden, collate fragmented data into clinically meaningful information, support decision-making, and extend precision education to clinicians and patients. This editorial accompanies the JMIR Perioperative Medicine webinar held in May 2026 and presents our perspectives-informed by our clinical experience-on where AI can assist perioperative clinicians and surgical patients, the evidence available today, what is still needed, and how clinicians can prepare to practice alongside AI tools. We consider opportunities in preoperative risk assessment and optimization; intraoperative intelligent decision support; emerging technologies, such as surgical digital twins; and postoperative recovery, monitoring, and prediction of clinical deterioration. Most AI applications remain early and require large-scale, prospective validation across diverse populations and resource settings. Realizing their potential also requires confronting substantial challenges, including limited and fragmented training data, the specialized models needed to interpret the large volumes of data captured in operative videos and during physiological monitoring, biases embedded in data and algorithms or biases inherent in human-AI interactions, and the determination of where human oversight is essential and where it is not feasible. When used responsibly, with human judgment remaining paramount, AI offers an unprecedented opportunity to advance perioperative care.
SN - 2561-9128
ER -