Endoscopy 2022; 54(S 01): S27
DOI: 10.1055/s-0042-1744611
Abstracts | ESGE Days 2022
ESGE Days 2022 Oral presentations
13:30–14:30 Thursday, 28 April 2022 Club A. Esophageal early cancer: Is ESD ready to take on the challenge?

ENDOSCOPIST PERFORMANCE IN OPTICAL DIAGNOSIS OF COLORECTAL POLYPS IN ARTIFICIAL INTELLIGENCE STUDIES

S. Pecere
1   Digestive Endoscopy Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy
,
G. Antonelli
2   Gastroenterology and Digestive Endoscopy Unit, Ospedale dei Castelli Hospital, Rome, Italy, Rome, Italy
,
M. Dinis-Ribeiro
3   CIDES/CINTESIS, Faculty of Medicine, University of Porto, Porto, Portugal, Porto, Portugal
5   Department of Biomedical Sciences, Humanitas University; 5 Humanitas Clinical and Research Center – IRCCS, Milano, Italy
,
Y. Mori
4   Digestive Disease Center, Showa University Northern Yokohama Hospital, Yokohama, Japan, Yokohama, Japan
,
C. Hassan
5   Department of Biomedical Sciences, Humanitas University; 5 Humanitas Clinical and Research Center – IRCCS, Milano, Italy
,
L. Fuccio
6   Department of Medical and Surgical Sciences (DIMEC), University of Bologna, S. Orsola-Malpighi Hospital, Bologna, Italy, Bologna, Italy
,
R. Bisschops
7   University Hospitals Leuven, Department of Gastroenterology and Hepatology, TARGID, KU Leuven, Belgium, Leuven, Belgium
,
G. Costamagna
1   Digestive Endoscopy Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy
,
E.H. Ji
8   Department of Internal Medicine, Healthcare Research Institute, Seoul National University Hospital Healthcare System Gangnam Center, Seoul, Korea, Seul, Korea, Republic of
,
D. Lee
8   Department of Internal Medicine, Healthcare Research Institute, Seoul National University Hospital Healthcare System Gangnam Center, Seoul, Korea, Seul, Korea, Republic of
,
M. Misawa
4   Digestive Disease Center, Showa University Northern Yokohama Hospital, Yokohama, Japan, Yokohama, Japan
,
H. Messmann
9   III Medizinische Klinik, Universitatsklinikum Augsburg, Augsburg, Germany
,
F. Iacopini
2   Gastroenterology and Digestive Endoscopy Unit, Ospedale dei Castelli Hospital, Rome, Italy, Rome, Italy
,
L. Petruzziello
1   Digestive Endoscopy Unit, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy
,
A. Repici
10   Department of Biomedical Sciences, Humanitas University; 11 Humanitas Clinical and Research Center – IRCCS, Milan, Italy
,
Y. Saito
11   Endoscopy Division, National Cancer Center Hospital, Tokyo, Japan., Tokyo, Japan
,
P. Sharma
12   Department of Gastroenterology and Hepatology, University of Kansas Medical Center, Kansas City, KS, Kansas City, United States
,
M. Yamada
11   Endoscopy Division, National Cancer Center Hospital, Tokyo, Japan., Tokyo, Japan
,
C. Spada
13   Centre for Endoscopic Research Therapeutics and Training (CERTT), Università Cattolica del Sacro Cuore, Rome, Italy
,
L. Frazzoni
6   Department of Medical and Surgical Sciences (DIMEC), University of Bologna, S. Orsola-Malpighi Hospital, Bologna, Italy, Bologna, Italy
› Author Affiliations
› Further Information
Also available at
 

Aims Widespread adoption of optical diagnosis of colorectal neoplasia is prevented by suboptimal endoscopist performance and lack of standardized training and competence evaluation.

We aimed to assess diagnostic accuracy of endoscopists in optical diagnosis of colorectal neoplasia in the framework of artificial intelligence(AI) validation studies.

Zoom Image
Fig. 1

Methods Literature searches of databases(PubMed/MEDLINE, EMBASE, Scopus) up to April 2021 were performed to identify articles evaluating accuracy of individual endoscopists in performing optical diagnosis of colorectal neoplasia within studies validating AI against a histologically verified ground-truth. The main outcomes were endoscopists’ pooled sensitivity, specificity, positive and negative predictive value(PPV/NPV), positive and negative likelihood ratio(LR) and area under the curve(AUC for sROC) for predicting adenomas vs non-adenomas.

Results Six studies with 67 endoscopists and 2085(IQR: 115-243,5) patients were evaluated. Pooled sensitivity and specificity for adenomatous histology was respectively 84.5% (95% CI 80.3-88%) and 83%(95% CI 79.6-85.9%), corresponding to a PPV, NPV, LR+, LR- of 89.5% (95% CI 87.1-91.5%), 75.7% (95% CI 70.1-80.7%), 5 (95% CI 3.9-6.2%) and 0.19 (95% CI 0.14-0.25%). The AUC was 0.82 (CI 0.76-0.90). Expert endoscopists showed a higher sensitivity than non-experts (90.5%, [95% CI 87.6-92.7%] vs. 75.5%, [95% CI 66.5-82.7%], p<0.001), and Eastern endoscopists showed a higher sensitivity than Western (85%, [95% CI 80.5-88.6%] vs. 75.8%, [95% CI 70.2-80.6%]). Quality was graded high for 3 studies and low for 3 studies.

Conclusions We show that human accuracy for diagnosis of colorectal neoplasia in the setting of AI studies is suboptimal. Educational interventions should especially target Western endoscopists and could benefit by AI validation settings.



Publication History

Article published online:
14 April 2022

© 2022. European Society of Gastrointestinal Endoscopy. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany