Comparison of Artificial Intelligence With Other Interventions to Improve Adenoma Detection Rate for Colonoscopy Article (Web of Science)

abstract

  • Introduction: Recent randomized controlled trials (RCTs) and meta-analysis have demonstrated improved adenoma detection rate (ADR) for colonoscopy with artificial intelligence (AI) compared with high-definition (HD) colonoscopy without AI. We aimed to perform a systematic review and network meta-analysis of all RCTs to assess the impact of AI compared with other endoscopic interventions aimed at increasing ADR such as distal attachment devices, dye-based/virtual chromoendoscopy, water-based techniques, and balloon-assisted devices. Methods: A comprehensive literature search of PubMed/Medline, Embase, and Cochrane was performed through May 6, 2022, to include RCTs comparing ADR for any endoscopic intervention mentioned above. Network meta-analysis was conducted using a frequentist approach and random effects model. Relative risk (RR) and 95% CI were calculated for proportional outcome. Results: A total of 94 RCTs with 61,172 patients (mean age 59.1±5.2 y, females 45.8%) and 20 discrete study interventions were included. Network meta-analysis demonstrated significantly improved ADR for AI compared with autofluorescence imaging (RR: 1.33, CI: 1.06 to 1.66), dye-based chromoendoscopy (RR: 1.22, CI: 1.06 to 1.40), endocap (RR: 1.32, CI: 1.17 to 1.50), endocuff (RR: 1.19, CI: 1.04 to 1.35), endocuff vision (RR: 1.26, CI: 1.13 to 1.41), endoring (RR: 1.30, CI: 1.10 to 1.52), flexible spectral imaging color enhancement (RR: 1.26, CI: 1.09 to 1.46), full-spectrum endoscopy (RR: 1.40, CI: 1.19 to 1.65), HD (RR: 1.41, CI: 1.28 to 1.54), linked color imaging (RR: 1.21, CI: 1.08 to 1.36), narrow band imaging (RR: 1.33, CI: 1.18 to 1.48), water exchange (RR: 1.22, CI: 1.06 to 1.42), and water immersion (RR: 1.47, CI: 1.19 to 1.82). Conclusions: AI demonstrated significantly improved ADR when compared with most endoscopic interventions. Future RCTs directly assessing these associations are encouraged.

authors

  • Aziz, Muhammad
  • Haghbin, Hossein
  • Sayeh, Wasef
  • Alfatlawi, Halah
  • Gangwani, Manesh K.
  • Sohail, Amir H.
  • Zahdeh, Tamer
  • Weissman, Simcha
  • Kamal, Faisal
  • Lee-Smith, Wade M
  • Nawras, Ali
  • Sharma, Prateek
  • Shaukat, Aasma

publication date

  • 2024

published in

number of pages

  • 12

start page

  • 143

end page

  • 155

volume

  • 58

issue

  • 2