RESUMEN

BACKGROUND AND AIMS: 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. METHODS: The MEDLINE database was searched through May 2020 for relevant articles by using key words such as machine learning, deep learning, artificial intelligence, computer-aided diagnosis, convolutional neural networks, GI endoscopy, and endoscopic image analysis. References and citations of the retrieved articles were also evaluated to identify pertinent studies. The manuscript was drafted by 2 authors and reviewed in person by members of the American Society for Gastrointestinal Endoscopy Technology Committee and subsequently by the American Society for Gastrointestinal Endoscopy Governing Board. RESULTS: Deep learning techniques such as convolutional neural networks have been used in several areas of GI endoscopy, including colorectal polyp detection and classification, 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. CONCLUSIONS: The implementation of AI technologies across multiple GI endoscopic applications has the potential to transform clinical practice favorably and improve the efficiency and accuracy of current diagnostic methods.

RESUMEN

Detecting and resecting colonic polyps during endoscopy has been shown to reduce CRC. At the same time, endoscopy has faced substantial technical improvements and continues to undergo major changes. Detection of colorectal neoplasias is facilitated by using high-definition endoscopy, and characterization of lesions is more accurate by using virtual CE. High-definition endoscopy has found widespread use, whereas virtual CE is still more often used in tertiary centers and for high-risk patient groups. Endomicroscopy has opened the door for in vivo histology and functional as well as molecular imaging. Thorough training of the endoscopists is necessary, and redefined reimbursement codes may help make these techniques available in the clinical arena for high-risk patients who are most probable to benefit from the more advanced diagnostic endoscopy techniques. Over recent years, there has also been an increasing push by healthcare payors and providers to improve the quality of healthcare services. The ultimate goal of monitoring quality measures is to improve the overall health of the population and, for colon cancer screening, to reduce the incidence and mortality of colon cancer. However, the development of an appropriate quality measure and ensuring the accuracy of the data on which we base our efforts are among the first challenges that we must meet (Table 1). The continued technologic advancements in colonoscopy and endoscopic imaging may result in improvements in the quality of colonoscopy and should lead to a decrease in the incidence and mortality from colon cancer.

Asunto(s)

RESUMEN

BACKGROUND: Traditional white-light endoscopy cannot reliably distinguish between small (<10 mm) adenomatous and hyperplastic colon polyps. High-definition white-light (HDWL) endoscopy and i-Scan may improve in vivo characterization of small colon polyps. OBJECTIVE: To compare HDWL endoscopy and HDWL plus i-Scan for the assessment of small colon polyps and to measure performance against the American Society for Gastrointestinal Endoscopy (ASGE) thresholds for assessment of diminutive colon polyps. DESIGN: Prospective cohort study. SETTING: Single academic hospital. PATIENTS: Patients undergoing bowel cancer screening colonoscopy. INTERVENTION: In vivo assessment of all polyps <10 mm by using HDWL and i-Scan image enhancement. MAIN OUTCOME MEASUREMENTS: The primary outcome measure was overall diagnostic accuracy of in vivo assessment of colon polyps <10 mm. Secondary outcome measures were sensitivity and specificity for adenomatous histology, negative predictive value for adenomatous histology of diminutive rectosigmoid polyps, and accuracy of prediction of polyp surveillance intervals. RESULTS: A total of 209 polyps in 84 patients were included. There were no significant differences between HDWL endoscopy and i-Scan in characterization of polyps <10 mm (accuracy 93.3% vs 94.7%; P = 1.00; sensitivity 95.5% vs 97.0%; P = .50; specificity 89.3% vs 90.7%; P = 1.00). The negative predictive value for adenomatous histology of diminutive rectosigmoid polyps was 100% with both HDWL endoscopy and i-Scan. U.K. and U.S. polyp surveillance intervals were predicted with 95.2% accuracy with HDWL endoscopy and 97.2% accuracy with i-Scan. LIMITATIONS: Single-center study. CONCLUSION: HDWL endoscopy may be as accurate as HDWL with i-Scan image enhancement for the in vivo characterization of small colon polyps. Both modalities fulfil the ASGE performance thresholds for the assessment of diminutive colon polyps. ( CLINICAL TRIAL REGISTRATION NUMBER: NCT01761279.).

Asunto(s)