Table 1.

Application of artificial intelligence (AI) in the patient safety of perioperative medication.

AI technology	Applications
Predictive analytics	AI algorithms have the potential to analyze vast amounts of data, identify patterns, and predict medication errors [28]. For example, clinicians can proactively intervene to prevent errors before they occur by analyzing medication orders and patient data to identify patients at high risk for medication errors, using AI.
Natural language processing	Natural language processing algorithms can analyze free-text notes in the electronic health record to identify potential medication errors that may have been missed through other means, such as notes that mention medication errors or adverse drug events [29]. This approach can assist clinicians in identifying and addressing potential errors before they cause harm.
Clinical decision support	AI can provide real-time decision support to clinicians by analyzing medication orders and providing alerts for potential drug interactions, dosing errors, or other safety concerns [30]. This feature can assist clinicians in making informed decisions about medication orders and reducing the risk of errors.
Machine learning	Machine learning algorithms can be used to identify patterns and predict medication errors by analyzing medication orders and patient data. These algorithms can also be used to develop personalized medication regimens for individual patients based on their unique characteristics, which can improve medication safety and reduce the risk of adverse drug events [31].
Computer vision	Computer vision algorithms can be used with barcoding systems to verify medication administration [22]. For instance, computer vision can analyze barcode scans to verify that the medication matches the order and the patient's information in the electronic health record. This feature can help reduce the risk of errors due to incorrect medication administration.