Skip to content

Artificial intelligence lesion Identification System based on dynamic big data of digestive endoscopy

Artificial intelligence lesion Identification System based on dynamic big data of digestive endoscopy

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2000039330
Enrollment
Unknown
Registered
2020-10-23
Start date
2020-10-23
Completion date
Unknown
Last updated
2021-01-25

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

Conditions

Digestive System Disease

Interventions

of&#32
and&#32
artificial&#32
Gold Standard:Pathological results
diagnosis&#32
system

Sponsors

Beijing Friendship Hospital, Capital Medical University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Selection criteria for artificial intelligence-assisted diagnosis of endoscopic images and videos: (1) The patient is between 18-80 years old and has not taken antiplatelet drugs such as aspirin or Plavix in the past week; (2) The patients endoscopic pictures and video information show that the gastrointestinal tract is well prepared and the visual field is well exposed; (3) The patient's endoscopic pictures and videos can accurately and clearly capture the location, size and shape of the lesion from multiple angles; (4) The patient is standardized by a senior endoscopy physician (the number of gastrointestinal endoscopy operations is more than 1,000); (5) The patients endoscopic pictures and video data must have pathological results judged as early cancer (case group) or non-early cancer (control group) as the gold standard support; 2. Selection criteria for artificial intelligence-assisted diagnosis of pathological images: (1) The patient's age is between 18-80 years old; (2) Patients with pathological diagnosis of early gastrointestinal cancer (WEO/Japan standard) are treated as the case group, and common non-gastrointestinal early cancers such as gastrointestinal tubular adenoma, ulcerative colitis and Crohn’s disease are used as control groups. Both the control group and the control group need HE staining or other staining pictures of pathological specimens; (3) The picture is clear and undamaged.

Exclusion criteria

Exclusion criteria: Nil

Design outcomes

Primary

MeasureTime frame
sensitivity;specificity;

Countries

China

Contacts

Public ContactShengtao Zhu

Beijing Friendship Hospital ,Capital Medical University

zhushengtao@ccmu.edu.cn+86 13911313236

Outcome results

None listed

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