Artificial Intelligence, Endoscopy, Gastrointestinal
Conditions
Brief summary
The objective of this study is to assess the effectiveness of an AI-based reporting system for upper gastrointestinal endoscopy. The primary question that this study aims to address is whether the reporting system can enhance the completeness and accuracy of endoscopic reports when assisted by AI, as drafted by endoscopists. Patients will be randomly assigned to either the experimental group or the control group. In the experimental group, physicians will draft EGD reports with the assistance of the AI-based reporting system, while in the control group, physicians will use the conventional reporting system to draft EGD reports. At the same time, the AI-based reporting system will automatically generate a report of the EGD examination.
Interventions
DIAGNOSTIC_TESTAI-based reporting system
AI-based reporting system is a software platform for real-time analysis and records of abnormalities and landmarks during endoscopy.
Sponsors
Renmin Hospital of Wuhan University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
DIAGNOSTIC
Masking
SINGLE (Subject)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
1. Aged 18 years 2. Aim to undergo screening, surveillance, and diagnosis 3. Undergo sedated EGD 4. Able to read, understand, and sign informed consent
Exclusion criteria
1. EGD contraindications 2. Not suitable for sedated endoscopy after anaesthesia evaluation 3. Biopsy contraindications 4. Active upper gastrointestinal bleeding or emergency oesophagogastroduodenoscopy (EGD) 5. Pregnancy 6. Upper gastrointestinal surgery or residual stomach 7. Not suitable for recruitment after investigator evaluation because of other high-risk conditions
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Completeness of reporting lesions | one month | Calculation method = number of report lesions / total number of lesions x 100% |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Accuracy of report drafting on lesion features | one month | Calculation method = number of accurately drafted features of lesions / total number of drafted features x 100% |
| Reporting time | one month | The time that endoscopists draft reports |
| Completeness of report drafting on lesion features | one month | Calculation method = number of drafted features of lesions / total number of features required to be drafted x 100% |
| Accuracy of report drafting on lesion features of AI system | one month | Calculation method = number of accurately drafted features of lesions / total number of drafted features x 100% |
| Physician satisfaction survey | one month | Use 5-point Likert scale to assess physician satisfaction, acceptance, and trust in using the intelligent graphic report system to draft endoscopic reports. |
| Completeness of reporting lesions of AI system | one month | Calculation method = number of report lesions / total number of lesions x 100% |
Contacts
Primary ContactHonggang Yu, MD
Outcome results
None listed