From discovering safer drugs to predicting complications before they happen, a new medical review maps how artificial intelligence (AI) is changing care in and around the operating room.  

What’s new

A mini‑review in Anesthesiology and Perioperative Science highlights four ways AI is reshaping anesthesiology and perioperative medicine: protein design, faster drug discovery, biomarker discovery, and day‑to‑day clinical practice. An overview diagram on page 2 distills these domains, showing how tools such as AlphaFold (which predicts a protein’s 3D shape) and modern machine‑learning models connect lab breakthroughs to bedside decisions.  

Why it matters

• Smarter, faster drug development. AI can spot promising drug targets, virtually screen compounds, predict side effects earlier, and even streamline clinical trials. A summary table on page 5 lays out AI’s roles—from target recognition to drug repositioning—along with the benefits (speed, lower cost) and challenges (data quality, false positives).  
• Personalized anesthesia. By analyzing large sets of health records and genetic data, AI can help tailor anesthesia plans—dose, drug choice, and monitoring—to each patient, aiming for safer procedures and smoother recoveries. The review describes how AI is being used to optimize real‑world use of the sedative dexmedetomidine, including modeling different dosing strategies and routes of administration.  
• Earlier warnings, fewer complications. In perioperative care, AI models can flag patients at risk for problems like acute kidney injury (AKI). One meta‑analysis summarized in the review reports AI systems reaching about 77% sensitivity and 75% specificity for predicting perioperative AKI—performance that could enable preventive steps before harm occurs.  

Mind the caveats

The authors stress that these systems aren’t magic. Protein predictions don’t always capture how drugs bind in real life, and clinical AI must be transparent and fair. Key hurdles include making models explainable for clinicians, building diverse datasets to avoid bias, protecting patient privacy, and fostering collaboration across disciplines.  

The bottom line

AI is moving from the lab to the operating room—guiding drug discovery, personalizing anesthesia, and helping clinicians anticipate complications. With stronger data, clearer ethics, and closer teamwork between computer scientists and clinicians, the review argues, AI could make anesthesia safer and more precise for everyone.  

Source: Fei Q., Zhang Y., Liu C., Zheng J., Fu Q. “Artificial intelligence in anesthesia and perioperative medicine,”Anesthesiology and Perioperative Science (2025).  

Editor’s note: This article is for information only and is not a substitute for professional medical advice.