Advertisement

Emerging role of artificial intelligence in GI endoscopy

      Artificial intelligence (AI) is a broad descriptor term that includes machine learning (ML) in which the algorithm, based on the input raw data, analyzes features in a separate dataset without specifically being programmed and delivers a specified classification (Fig. 1). Deep learning techniques such as convolutional neural networks (CNNs) are transformative ML techniques that enable rapid and accurate image discrimination and classification and as such have many applications within medicine. In gastroenterology, CNNs have been used in several areas of GI endoscopy, including colorectal polyp detection, and classification, including assessment of the presence of advanced neoplasia in colonic polyps, evaluation of histologic inflammation in endocytoscopic images obtained during colonoscopy in patients with ulcerative colitis, analysis of endoscopic images for diagnosis of Helicobacter pylori infection, detection and depth assessment of early gastric cancer, dysplasia in Barrett’s esophagus, and detection of various abnormalities in wireless capsule endoscopy images (Table 1). Although currently an AI system that can be used in clinical practice is not commercially available, the most promising initial application appears to be for real-time colonic polyp detection and classification. Critical appraisal of the improvement in patient outcomes, cost-effectiveness, and safety and the changes in clinical practice required to incorporate and implement these tools are required. Adopting these technologies will be associated with some cost burden; a corresponding reimbursement for their use will undoubtedly affect the rate of incorporation into clinical practice. The accompanying technology committee document (available online at https://doi.org/10.1016/j.vgie.2020.08.013) describes in detail the currently reported applications of AI in gastroenterology, focusing on endoscopic image analysis.
      • Pannala R.
      • Krishnan K.
      • Melson J.
      • et al.
      Artificial intelligence in gastrointestinal endoscopy.
      Figure thumbnail gr1
      Figure 1Diagrammatic representation of hierarchy of AI domains. (Adapted with permission from Goodfellow I, Bengio Y, Courville A. Deep learning. Cambridge, MA: MIT Press; 2016.). AI, Artificial intelligence; ML, machine learning; RL, representation learning; DL, deep learning.
      Table 1Reported applications of computer-aided diagnosis and artificial intelligence in various endoscopic procedures
      Procedure Application
      Colonoscopy Detection of polyps (real-time and on still images and video)
      Applications where use of deep learning has been reported.
      ; CADe

      Classification of polyps (neoplastic vs hyperplastic)
      Applications where use of deep learning has been reported.
      ; CADx

      Detection of malignancy within polyps (depth of invasion on endocytoscopic images)
      Applications where use of deep learning has been reported.


      Presence of mucosal inflammation on endocytoscopic images
      Applications where use of deep learning has been reported.


      Assessment of disease activity in inflammatory bowel disease
      Applications where use of deep learning has been reported.


      Assessment of quality metrics in colonoscopy
      Wireless capsule endoscopy Lesion detection and classification (bleeding, ulcers, polyps)
      Applications where use of deep learning has been reported.


      Assessment of intestinal motility

      Celiac disease (assessment of villous atrophy, intestinal motility)

      Improve efficiency of image review
      • Deletion of duplicate images and uninformative image frames (eg, images with debris)
        Applications where use of deep learning has been reported.
      Upper endoscopy Identify anatomic location
      Applications where use of deep learning has been reported.


      Diagnosis of Helicobacter pylori infection
      Applications where use of deep learning has been reported.


      Gastric cancer detection and assessing depth of invasion
      Applications where use of deep learning has been reported.


      Esophageal squamous dysplasia

      Detection and delineation of early dysplasia in Barrett’s esophagus
      Applications where use of deep learning has been reported.


      Real-time image segmentation in volumetric laser endomicroscopy in Barrett’s esophagus
      Applications where use of deep learning has been reported.
      EUS Differentiation of pancreatic cancer from chronic pancreatitis and normal pancreas

      Differentiation of autoimmune pancreatitis from chronic pancreatitis

      EUS elastography
      CADe, Computer-aided detection; CADx, computer-aided diagnosis.
      Applications where use of deep learning has been reported.
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Gastrointestinal Endoscopy
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      Reference

        • Pannala R.
        • Krishnan K.
        • Melson J.
        • et al.
        Artificial intelligence in gastrointestinal endoscopy.
        VideoGIE. Epub 2020 Nov 9;

      Linked Article

      • Artificial intelligence in gastrointestinal endoscopy
        VideoGIEVol. 5Issue 12
        • Preview
          Artificial intelligence (AI)-based applications have transformed several industries and are widely used in various consumer products and services. In medicine, AI is primarily being used for image classification and natural language processing and has great potential to affect image-based specialties such as radiology, pathology, and gastroenterology (GE). This document reviews the reported applications of AI in GE, focusing on endoscopic image analysis.
        • Full-Text
        • PDF
        Open Access