Artificial Intelligence for Determination of Gastroscopy Surveillance Intervals

Gastric Cancer Clinical Trial

Official title:

Development and Validation of Gastroscopy Surveillance Recommendations Based on Natural Language Processing for Patients With Gastric Cancer and Precancerous Diseases

Clinical Trial Details — Status: Active, not recruiting

Administrative data

• NCT number: NCT05631015
• Other study ID #: 2022-SDU-QILU-G008
• Status: Active, not recruiting
• Start date: January 1, 2012
• Est. completion date: December 31, 2023

Study information

• Verified date: November 2022
• Source: Shandong University
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

The purpose of this study is to develop and validate a clinical decision support system based on automated algorithms. This system can use natural language processing to extract data from patients' endoscopic reports and pathological reports, identify patients' disease types and grades, and generate guidelines based follow-up or treatment recommendations

Recruitment information / eligibility

• Status: Active, not recruiting
• Enrollment: 2000
• Est. completion date: December 31, 2023
• Est. primary completion date: October 31, 2022
• Gender: All
• Age group: 18 Years to 80 Years

Eligibility

Inclusion Criteria:

• Patients aged 18 - 80 years

• Patients underwent endoscopic examination

Exclusion Criteria:

• Patients with the contraindications to endoscopic examination

• Patients with imcomplete examination information

• Patients undergo endoscopy for therapy

• Patients have history of upper gastrointestinal surgery

• Patients with duodenal or Laryngeal neoplasms

• Patients with gastrointestinal submucosal tumor

Related Conditions & MeSH terms

• Atrophic Gastritis
• Carcinoma in Situ
• Early Gastric Cancer
• Gastric Cancer
• Gastritis
• Gastritis, Atrophic
• Helicobacter Pylori Infection
• High Grade Intraepithelial Neoplasia
• Intestinal Metaplasia
• Metaplasia
• Neoplasms
• Stomach Neoplasms

Intervention

Other:
• AI recongnize disease and generate recommendations
According the endoscopic reports and pathological reports, the decision support system recognise patients' disease types and grades, and generate guidelines based survilliance or treatment recommendations.

Locations

Country	Name	City	State
China	Qilu Hospital, Shandong University	Jinan	Shandong

Sponsors (1)

Lead Sponsor	Collaborator
Xiuli Zuo

Country where clinical trial is conducted

China, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	The diagnostic accuracy of gastric diseases with deep learning algorithm	The diagnostic accuracy of gastric diseases with deep learning algorithm	12 month
Primary	The accuracy of recommentions for different disease with deep learning algorithm	The accuracy of recommentions for different disease with deep learning algorithm	12 month
Secondary	The diagnostic sensitivity of gastric diseases with deep learning algorithm	The diagnostic sensitivity of gastric diseases with deep learning algorithm	12 month
Secondary	The diagnostic specificity of gastric diseases with deep learning algorithm	The diagnostic specificity of gastric diseases with deep learning algorithm	12 month
Secondary	The diagnostic positive predictive value of gastric diseases with deep learning algorithm	The diagnostic positive predictive valu of gastric diseases with deep learning algorithm	12 month
Secondary	The diagnostic negative predictive value of gastric diseases with deep learning algorithm	The diagnostic negative predictive value of gastric diseases with deep learning algorithm	12 month
Secondary	The F-score of gastric diseases with deep learning algorithm	The F-score of gastric diseases with deep learning algorithm	12 month