Artificial intelligence in endoscopy: Overview, applications, and future directions

Abstract

Since the emergence of artificial intelligence (AI) in medicine, endoscopy applications in gastroenterology have been at the forefront of innovations. The ever-increasing number of studies necessitates the need to organize and classify applications in a useful way. Separating AI capabilities by computer aided detection (CADe), diagnosis (CADx), and quality assessment (CADq) allows for a systematic evaluation of each application. CADe studies have shown promise in accurate detection of esophageal, gastric and colonic neoplasia as well as identifying sources of bleeding and Crohn's disease in the small bowel. While more advanced CADx applications employ optical biopsies to give further information to characterize neoplasia and grade inflammatory disease, diverse CADq applications ensure quality and increase the efficiency of procedures. Future applications show promise in advanced therapeutic modalities and integrated systems that provide multimodal capabilities. AI is set to revolutionize clinical decision making and performance of endoscopy.

Keywords: Artificial intelligence; Barrett's; IBD; capsule endoscopy; endoscopy; polyp.

Figure 1

Hierarchy of AI domains

Figure 2

A framework for characterizing applications of AI in endoscopy

Figure 3

Example of CADe and CADx application in colonic polyps. Top image shows CADe with a blue bounding box “detecting” a polyp. The bottom image shows a different colored bounding box where an “optical biopsy” has been performed by the CADx algorithm. (Image courtesy of Michael Byrne on behalf of Satisfai Health)

Figure 4

CADx detection of dysplasia and esophageal adenocarcinoma in Barrett’s esophagus.[22] A segmented overlay highlights the area of suspected dysplasia in the endoscopy output. (Image courtesy of Michael Byrne on behalf of Satisfai Health)

Figure 5

Automated detection and grading of severity in ulcerative colitis.[66] The graphical interface of the AI output provides clinical disease scoring (right panel) and identifies areas of disease activity within the endoscopy output (lower bar), directing reviewers to these for confirmation. (Image courtesy of Michael Byrne on behalf of Satisfai Health)

Figure 6

Major applications of artificial intelligence in endoscopy within GI compartments