The study includes patients with common stomach diseases and healthy controls, with diseases including gastroesophageal reflux, various types of gastritis, and gastric ulcers
Conditions
Interventions
Lesion detection and classification model group:None
System Integration and Verification Team:None
Report generation module group:None
Sponsors
The First Affiliated Hospital of Dalian Medical University
Eligibility
Sex/Gender
All
Age
18 Years to No maximum
Inclusion criteria
Inclusion criteria: 1. Age >= 18 years old; 2. From 2023 to 2025, he had gastroscopies at various digestive endoscopy centres, the endoscopy videos were clear and the clinical reports were complete and standardised; 3. Clinical data is complete, with no factors affecting the assessment of lesions.
Exclusion criteria
Exclusion criteria: 1. Poor quality of endoscopic videos (such as halo, blurriness, excessive mucus or blood content, etc.); 2. Combined severe gastrointestinal malformations, advanced malignant tumours, or other serious organic diseases; 3. Has a history of upper gastrointestinal surgery; 4. Clinical report information is missing or not standardised, so it can't be used for model validation.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Report Generation Accuracy;Lesion Detection Sensitivity;Lesion Detection Specificity; | — |
Secondary
| Measure | Time frame |
|---|---|
| Agreement between Model and Endoscopist Reports (Kappa Coefficient);Report Generation Efficiency;Explainability Metrics (w-Precision, IAUC, etc.); | — |
Countries
China
Contacts
Public ContactDuan Zhijun
The First Affiliated Hospital of Dalian Medical University
Outcome results
None listed