Endoscopy
DOI: 10.1055/a-2626-0069
Original article

A novel artificial intelligence-based system for quality monitoring during esophagogastroduodenoscopy: a multicenter randomized controlled study

Shengsen Chen
1   Department of Endoscopy, Zhejiang Cancer Hospital, Hangzhou, China (Ringgold ID: RIN89680)
2   Graduate Student Training Base Alliance, Wenzhou Medical University, Wenzhou, China
,
Yandong Li
1   Department of Endoscopy, Zhejiang Cancer Hospital, Hangzhou, China (Ringgold ID: RIN89680)
,
Huogen Wang
3   Hithink RoyalFlush Information Network Co. Ltd., Hangzhou, China
,
Ming Chen
3   Hithink RoyalFlush Information Network Co. Ltd., Hangzhou, China
,
Chun Wang
3   Hithink RoyalFlush Information Network Co. Ltd., Hangzhou, China
,
Weilong Pan
4   Department of Gastroenterology, Xinchang People’s Hospital, Shaoxing, China
,
Ji Ma
5   Department of Gastroenterology, The First People’s Hospital of Yongkang, Jinhua, China
,
Tingting Ji
6   Department of Gastroenterology, Ruian People’s Hospital, Wenzhou, China
,
Jiemin Hong
7   Department of Gastroenterology, The Second Hospital of Yinzhou, Ningbo, China
,
Di Hu
8   Department of Gastroenterology, The Ninth Hospital of Ningbo, Ningbo, China
,
Xiancang Sheng
9   Department of Gastroenterology, The First People’s Hospital of Taizhou, Taizhou, China
,
Junbing Jiang
10   Department of Gastroenterology, The First People’s Hospital of Linhai, Taizhou, China
,
1   Department of Endoscopy, Zhejiang Cancer Hospital, Hangzhou, China (Ringgold ID: RIN89680)
2   Graduate Student Training Base Alliance, Wenzhou Medical University, Wenzhou, China
› Author Affiliations

Supported by: Zhejiang Traditional Chinese Medicine Administration 2024ZL312
Supported by: Natural Science Foundation for Young Scientists 82402384
Supported by: Natural Science Foundation of Zhejiang Province LHDMY24H190002
Supported by: Medical and Health Research Project of Zhejiang Province 2024KY040 Clinical Trial: Registration number (trial ID): ChiCTR2100045597, Trial registry: Chinese Clinical Trial Registry (http://www.chictr.org/), Type of Study: Prospective, Randomized, Multi-Center Study


Preview

Abstract

Background

Esophagogastroduodenoscopy (EGD) is the pivotal procedure for diagnosis of upper gastrointestinal (UGI) lesions. However, significant variation in EGD performance among endoscopists impacts detection rates of UGI cancers and precursor lesions. We developed a novel EGD quality monitoring system and evaluated its effectiveness in a randomized controlled study.

Methods

The endoscopy quality control assistant (EQCA) was developed using deep convolutional neural networks and long short-term memory. Patients (≥18 years) undergoing EGD in seven hospitals were consecutively enrolled and randomly assigned to the EQCA-assisted group or control group. The primary outcome was the detection rate for cancer-related lesions (low and high grade intraepithelial neoplasia and cancer) and cancer (early and advanced cancer) in the UGI tract.

Results

After randomization and exclusions, 16 005 patients in the control group and 16 012 in the EQCA group were analyzed. Detection rates for UGI cancer-related lesions and cancer were significantly higher in the EQCA group than in the control group (8.00% vs. 5.55%; 1.93% vs. 1.21%; both P < 0.001). The EQCA group had a higher operation score, reflecting examination quality, and longer inspection time than the control group. The detection rate for UGI cancer-related lesions was positively correlated with operation score (r = 0.9217, P < 0.001) and inspection time (r = 0.8943, P < 0.001) for each hospital.

Conclusions

The use of EQCA during EGD was associated with increased detection of UGI cancer and precancerous lesions. Our novel EQCA system can be an effective tool for monitoring real-time EGD quality.

Supplementary Material



Publication History

Received: 12 July 2024

Accepted after revision: 23 May 2025

Article published online:
01 July 2025

© 2025. Thieme. All rights reserved.

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany