AI-assisted White Light Endoscopy to Identify the Kimura-Takemoto Classification of Atrophic Gastritis

Shandong University1,500 enrolled

Overview

Grading endoscopic atrophy according to the Kimura-Takemoto classification can assess the risk of gastric neoplasia development. However, the false negative rate of chronic atrophic gastritis is high due to the varying diagnostic standardization and diagnostic experience and levels of endoscopists. Therefore, this study aims to develop an AI model to identify the Kimura-Takemoto classification.

Grading endoscopic atrophy according to the Kimura-Takemoto classification can assess the risk of gastric neoplasia development. The higher the score, the more severe the degree of atrophic gastritis. However, the false negative rate of chronic atrophic gastritis is high due to the varying diagnostic standardization and diagnostic experience and levels of endoscopists. Therefore, this study aims to develop an AI model to identify the Kimura-Takemoto classification of atrophic gastritis to achieve gastric cancer risk assessment.

Study Type

OBSERVATIONAL

Enrollment

1,500

Conditions

Atrophic Gastritis Artificial Intelligence Endoscopy

Interventions

Diagnostic Test: The diagnosis of Artificial Intelligence and endosopistsDIAGNOSTIC_TEST

Endosopists and AI will assess the Kimura-Takemoto classification independently when the patients is eligible.

Eligibility

Sex: ALLMin age: 18 YearsMax age: 80 Years

Medical Language ↔ Plain English

Inclusion Criteria: Patients aged 18-80 years who undergo the white light endoscope examination Informed consent form provided by the patient. Exclusion Criteria: 1. patients with severe cardiac, cerebral, pulmonary or renal dysfunction or psychiatric; 2. disorders who cannot participate in gastroscopy; 3. Patients with progressive gastric cancer; 4. low quality pictures; 5. patients with previous surgical procedures on the stomach or esophageal; 6. patients who refuse to sign the informed consent form;

Locations (1)

Department of Gastrology, QiLu Hospital, Shandong University

Shangdong, Shandong, China

RECRUITING

Outcomes

Primary Outcomes

Accuracy of AI model to diagnose the Kimura-Takemoto classification

Time frame: 2 years

Sensitivity of AI model to diagnose the Kimura-Takemoto classification

Time frame: 2 years

Specificity of AI model to diagnose the Kimura-Takemoto classification

Time frame: 2 years

Secondary Outcomes

The MIOU value of AI model in semantic segmentation of endoscopic atrophy picture

Time frame: 2 years

Central Contacts

yanqing Li, MD, PHD

CONTACT

0531182169385 liyanqing@sdu.edu.cn

Data from ClinicalTrials.gov

This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.