Skip to content

Clinical validation of an artificial intelligence model for endoscopic classification of chronic atrophic gastritis

Clinical validation of an artificial intelligence model for endoscopic classification of chronic atrophic gastritis

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2400082633
Enrollment
Unknown
Registered
2024-04-02
Start date
2024-04-10
Completion date
Unknown
Last updated
2024-04-08

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

Conditions

Chronic atrophic gastritis

Interventions

Gold Standard:The Kimura-Takemoto classification is endoscopic image diagnosis, and the project team has established a diagnostic expert group consisting of three people (expert group: those who have
Index test:The AI model for identifying CAG and making the diagnosis of Kimura-Takemoto classification. Beginners: able to independently and standardly complete gastroscopy operations, with less than
Senior endoscopic experts: able to independently and standardly complete gastroscopy operations, with more than 10000 cases.

Sponsors

The First Affiliated Hospital of Army Medical University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 80 Years

Inclusion criteria

Inclusion criteria: Inclusion criteria for cases needed to verify the clinical efficacy of the AI model: ?patients who have completed painless electronic gastroscopy (the criteria are defined as those who have completed painless electronic gastroscopy because of negative verification) ?patients aged 18 years or older and signed informed consent ?complete intraluminal pictures or video data are kept in the endoscopic system ?endoscopic doctors can accurately identify the location of the patient's image and accurately judge whether there is gastric mucosal atrophy and adequate exposure of gastric mucosal folds Inclusion criteria for endoscopic doctors needed to verify the clinical efficacy of the AI model: ?Beginners: able to independently and standardly complete gastroscopy operations, with less than 1000 cases ?Senior endoscopic experts: able to independently and standardly complete gastroscopy operations, with more than 10000 cases

Exclusion criteria

Exclusion criteria: ?Those who did not successfully complete painless electronic gastroscopy ?Those who were under the age of 18 or did not sign the informed consent ?Intraluminal images or video data not available in the endoscopic system ?The collected images were not clear enough, with blurring ?The collected images could not be accurately judged by the endoscopic doctors on the location of the shooting and whether there was atrophy

Design outcomes

Primary

MeasureTime frame
Accuracy rate;

Countries

China

Contacts

Public ContactChen Lei

The First Affiliated Hospital of Army Medical University

chenlei1977603@126.com+86 132 2868 3896

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026