Skip to content

A Prospective Study on the Utility of an Artificial Intelligence-Based Detection System for Esophageal and Gastric Tumors

A Prospective Non-Inferiority Trial Comparing AI-Assisted Conventional Endoscopy with Expert Magnifying Endoscopy for the Detection of Esophageal and Gastric Neoplasms - AIM-UGI Study

Status
Recruiting
Phases
Unknown
Study type
Interventional
Source
JPRN
Registry ID
JPRN-UMIN000060929
Enrollment
138
Registered
2026-03-20
Start date
2026-04-01
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

Esophageal Cancer, Gastric Cancer, Gastric adenoma

Interventions

Patients with confirmed esophageal or gastric cancer will undergo endoscopic examination using a conventional esophagogastroduodenoscopy (EG-860R) combined with an AI-based image diagnostic support sy

Sponsors

The Jikei University School of Medicine
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: Patients with a current diagnosis of esophageal cancer or gastric cancer (including adenoma) who are scheduled to undergo diagnostic esophagogastroduodenoscopy at Jikei University Kashiwa Hospital or Jikei University Katsushika Medical Center. Patients who have received a full explanation of the study, have demonstrated sufficient understanding, and have provided written informed consent of their own free will to participate in the study.

Exclusion criteria

Exclusion criteria: Patients who are minors (under 20 years of age). Patients who have received chemotherapy for a malignant tumor within the past 12 months. Patients in whom biopsy cannot be performed at the time of endoscopy (e.g., due to ongoing antithrombotic therapy). Patients in whom iodine staining of the esophagus cannot be performed due to allergy or other contraindications. Patients who are deemed inappropriate for enrollment in this study by the treating physician or examining endoscopist for any other reason.

Design outcomes

Primary

MeasureTime frame
Sensitivity for detecting esophageal and gastric neoplasms in the screening endoscopist group using a standard upper gastrointestinal endoscope with AI assistance compared with the expert endoscopist group using a magnifying endoscope (Sensitivity was defined as the number of esophageal squamous cell carcinomas, gastric adenomas, and gastric adenocarcinomas detected in each examination divided by the total number of these lesions identified by the referring physician and during the first and second study examinations).

Secondary

MeasureTime frame
Diagnostic accuracy for esophageal and gastric cancers (discriminative performance [sensitivity, specificity, positive predictive value, and negative predictive value], and accuracy of depth-of-invasion assessment). Analysis of lesions that were difficult to detect with each endoscopy (including lesion location, tumor type, and macroscopic size).

Countries

Japan

Contacts

Public ContactJumpei Yamamoto

The Jikei University School of Medicine Department of Endoscopy

juno178.7328@gmail.com+81-4-7164-1111

Outcome results

None listed

Source: JPRN (via WHO ICTRP) · Data processed: Jul 3, 2026