Article Text

Download PDFPDF

Highlights from the literature
  1. Julia Louisa Gauci1,
  2. Ian D Penman2
  1. 1 Gastrointestinal Unit, Western General Hospital, Edinburgh, UK
  2. 2 Centre for Liver and Digestive Disorders, Royal Infirmary of Edinburgh, Edinburgh, UK
  1. Correspondence to Dr Julia Louisa Gauci, Gastrointestinal Unit, Western General Hospital, Edinburgh, UK; julia.gauci{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Artificial intelligence (AI) improves the quality of diagnostic colonoscopy

Deep learning systems allow real-time computer-aided polyp detection (CADe). Early reports suggested CADe improves polyp detection, thus potentially reducing interval cancers. Hassan et al 1 published a systematic review and meta-analysis of five randomised controlled trials comparing adenoma detection rate (ADR) and the features of detected lesions when using CADe versus white light endoscopy (WLE) alone. A total of 4354 participants were included in the analysis (2163 for CADe and 219 for WLE). The overall ADR was significantly higher for the CADe group versus controls (791/2163 (36.6%) vs 558/2191 (25.2%), RR 1.44), with similar findings for the number of adenomas per colonoscopy (1249/2163 (0.58%) vs 779/2191 (0.36%), RR 1.70), irrespective of size, location or morphology. No statistically significant difference between withdrawal times in the two groups was observed. Incorporating CADe into diagnostic colonoscopy significantly increases the detection of colorectal neoplasia, regardless of adenoma characteristics, with no negative impact on efficiency. We may expect …

View Full Text


  • Twitter @JuGauc, @GastronautIan

  • Contributors Both authors contributed equally to selecting articles and preparing and reviewing the manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.