Perceptions of Artificial Intelligence in Endoscopy among Italian Gastroenterologists: a National Survey

M Maida; S Sferrazza; G Calabrese; G Marasco; M Furnari; I Boskoski; A Vitello; E Sinagra; A Facciorusso

doi:10.1055/s-0045-1805952

Endoscopy, Table of Contents

Endoscopy 2025; 57(S 02): S378
DOI: 10.1055/s-0045-1805952

Abstracts | ESGE Days 2025

ePosters

Perceptions of Artificial Intelligence in Endoscopy among Italian Gastroenterologists: a National Survey

Authors

M Maida

¹Department of Medicine and Surgery, University of Enna ‘Kore’, Enna, Italy

²Gastroenterology Unit, Umberto I Hospital, Enna, Italy
S Sferrazza

³Gastroenterology and Endoscopy Unit, 'ARNAS Civico – Di Cristina – Benfratelli' Hospital, Palermo, Italy
G Calabrese

³Gastroenterology and Endoscopy Unit, 'ARNAS Civico – Di Cristina – Benfratelli' Hospital, Palermo, Italy
G Marasco

⁴Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy

⁵S. Orsola-Malpighi Polyclinic, Bologna, Italy
M Furnari

⁶Division of Gastroenterology, Department of Internal Medicine, University of Genoa, Genoa, Italy

⁷IRCCS Ospedale Policlinico San Martino, Genoa, Italy
I Boskoski

⁸Digestive Endoscopy Unit, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy
A Vitello

¹Department of Medicine and Surgery, University of Enna ‘Kore’, Enna, Italy
E Sinagra

⁹Gastroenterology Unit, Fondazione Istituto San Raffaele, Cefalù, Italy
A Facciorusso

¹⁰Gastroenterology Unit, Department of Medical and Surgical Sciences, University of Foggia, Foggia, Italy

Abstract

Full Text

Aims An extensive body of evidence regarding artificial intelligence (AI) in endoscopy is currently available, but the transition of these technologies to the bedside is still in progress. This national survey aims to assess the perceptions of AI among gastroenterologists in Italy.

Methods 320 Italian gastroenterologists and trainees in gastroenterology were invited to answer a web-based survey.

Results Data from 150 respondents from 20 Italian regions were analyzed. Overall, 67.3% of the participants were practicing gastroenterologists, and 32.7% were trainees. 99.3% have heard of AI in gastroenterology and 49.3% of them currently use AI tools in endoscopy. Median perception of AI systems was 8 (6 – 8) for simplicity in use, 8 (3–7) for diagnostic sensitivity, 6 (5–8) for diagnostic specificity, and 5 (2–6) for extension of procedure time on a scale from 1 to 10. About one-half of participants raised a fair concern about using AI systems. The main concern was on regulatory issues, with a median score of 6 (4–7) on a scale from 1 to 10, followed by legal and data protection, with a median score of 5 (3–7) and 5 (2–7), respectively. On the contrary, fair concern was expressed regarding the reliability of the tools and ethical issues, with a median score of 4 (2–6). Moreover, 78.3% of respondents believe that AI should be applied in the endoscopy training of young gastroenterologists. At subgroup analysis, male gastroenterologists showed higher understanding of AI compared to female, as older gastroenterologists compared to younger ones, and gastroenterologists compared to trainees. Additionally, older gastroenterologists reported finding AI endoscopy tools simpler to use and expressed more concerns about regulatory issues.

Conclusions The data from this survey show that Italian gastroenterologists have proper awareness and favorable perception of AI in endoscopy, with a good diffusion of AI tools across the national territory and fair concerns about regulatory issues.