Anaesthesiologists have an ever-increasing number of tools at their disposal to assess the risk that a patient will assume when undergoing anaesthesia and surgery, and the likelihood of specific events during surgery.

The earliest versions of these tools were based on heuristics or on a small number of easily identifiable features; the Revised Cardiac Risk Index (RCRI) and pulse pressure variation are good examples of this type of preoperative and intraoperative assessment tool.

However, as medicine and technology have evolved, extremely large amounts of healthcare data have become available as computation and data storage capabilities have grown ever cheaper.

This abundance of data and computational power has yielded the ability to calculate specific risk scores based on an increasing number of features, including risk calculators based on the National Surgical Quality Improvement Program (NSQIP) and Society of Thoracic Surgeons (STS) databases.

In the context of the oft-quoted limitation on human cognitive capacity to 5–10 facts per decision, researchers have argued for the use of artificial intelligence (AI)/machine learning (ML) techniques to tap the full potential of this flood of data to achieve truly personalised medical decision support.