A randomized controlled trial for reducing miss of esophageal and gastric neoplasms by using artificial intelligence

A multicenter randomized controlled trial for reducing miss of esophageal and gastric neoplasms by using artificial intelligence CAD-EYE - REGO trial

Status

Recruiting

Phases

Phase 3

Study type

Interventional

Source

JPRN

Registry ID

JPRN-jRCTs022240045

Enrollment

1600

Registered

2024-12-17

Start date

2025-04-03

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 neoplasm, gastric neoplasm esophageal neoplasm, gastric neoplasm

Interventions

In the AI-first group, AI-assisted endoscopic observation (blue light imaging [BLI] in esophagus and white light imaging [WLI] in stomach), followed by routine endoscopic observation (WLI in stomach a

D001185

Artificial intelligence

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: (1) Patients with high risk for gastric cancer (previous gastric cancer with endoscopic resection) or esophageal cancer (previous esophageal cancer or head and neck cancer treated with endoscopic resection or radiation therapy, and present head and neck cancer) (2) Patients with performance status 0-2 (3) The patients aged 18 years or more

Exclusion criteria

Exclusion criteria: (1) History of surgical resection of esophagus or stomach including resection of pharynx with esophageal resection (2) Contraindication to endoscopic biopsy (3) Contraindication to endoscopy (4) Pregnancy, 28 days or less postpartum, or breastfeeding during recruitment (5) Severe mental disease (6) Refusal to participate in this study

Design outcomes

Primary

Measure	Time frame
A miss rate of upper gastrointestinal (GI) neoplasm in the primary mode	—

Secondary

Measure	Time frame
A detection rate of upper GI neoplasm in the primary mode Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of upper GI neoplasm in the primary mode The occurrence rate of adverse events Miss and detection rates of each of esophageal and gastric neoplasms in the primary mode Sensitivity, specificity, PPV, and NPV of each of esophageal and gastric neoplasms in the primary mode Miss and detection rates, sensitivity, specificity, PPV, and NPV of upper GI cancer and each of esophageal and gastric cancers in the primary mode Observation time in each protocol examination A rate of endoscopic diagnosis for upper GI neoplasm in the primary mode Agreement of each endoscopic finding (macroscopic type, tumor size, and the color of the tumor) between examiners and a central reviewer	—

Measure

Time frame

A detection rate of upper GI neoplasm in the primary mode Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of upper GI neoplasm in the primary mode The occurrence rate of adverse events Miss and detection rates of each of esophageal and gastric neoplasms in the primary mode Sensitivity, specificity, PPV, and NPV of each of esophageal and gastric neoplasms in the primary mode Miss and detection rates, sensitivity, specificity, PPV, and NPV of upper GI cancer and each of esophageal and gastric cancers in the primary mode Observation time in each protocol examination A rate of endoscopic diagnosis for upper GI neoplasm in the primary mode Agreement of each endoscopic finding (macroscopic type, tumor size, and the color of the tumor) between examiners and a central reviewer

—

Contacts

Public ContactWaku Hatta

Tohoku University Graduate School of Medicine

waku-style@festa.ocn.ne.jp+81-22-717-7171

Outcome results

None listed

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