Artificial Intelligence for Smart Procedural Sedation in the Gastrointestinal Endoscopy Suite

Abstract

Artificial intelligence (AI) is defined as the science of creating intelligent machines. AI has grown exponentially, and its systems have made their way into the anesthesia field. The purpose of this review is to explore how the practice of anesthesiology in the gastrointestinal (GI) endoscopy suite changed with AI. Current AI anesthesia systems in the endoscopy suite include open and closed loop anesthesia delivery systems. The most widely used open loop system is the target-controlled infusion (TCI). During TCI, a drug is given automatically using a pump controlled by a computer. The aim is to achieve a chosen target plasma concentration, based on the hypothesis that the pharmacological effect is proportional to the drug’s plasma concentration. Closed loop systems regulate the drug’s dosage by checking a controlling parameter such as the patient himself in patient-maintained sedation, or the bispectral index in computer-assisted personalized sedation. As such, the closed loop system regulates the dose according to continuous feedback from the patient. Recent innovations in AI include machine learning and deep learning models that may have future applications in the endoscopy suite. Machine learning models look for patterns in vast amounts of data to draw conclusions. Deep learning models gain the ability to learn new information that they were not “explicitly programmed” to learn and make changes to their function based on that new information. Although the future of AI in anesthesia and the GI endoscopy suite seems bright, one must always keep in mind its shortcomings.