Development of an Artificial Intelligence System for Automatic Recognition of Anatomical Sites During Esophagogastroduodenoscopy

Research and development of AI for landmark recognition in Esophagogastroduodenoscopy - CAQ-EGD

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000061816

Enrollment

400

Registered

2026-06-15

Start date

2024-10-15

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

gastric cancer

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: Patients 18 years of age or older who underwent EGD at Showa Medical University Northern Yokohama Hospital between January 1, 2020 and October 31, 2025, who did not opt out of the use of their AI-related information.

Exclusion criteria

Exclusion criteria: (1) Incomplete EGD; (2) limited visualization due to gastric residue; (3) history of esophageal, gastric, or duodenal surgery.

Design outcomes

Primary

Measure	Time frame
Classification accuracy of the AI for standard anatomical landmarks of the esophagus, stomach, and duodenum.	—

Secondary

Measure	Time frame
Performance of the image-quality assessment AI for filtering low-quality images; per-site classification accuracy.	—

Countries

Japan

Contacts

Public ContactTomoya Shibuya

SHOWA Medical University Northern Yokohama Hospital Digestive Disease Center

tswork0603@gmail.com0459497000

Outcome results