In this study, we proposed a prospective study about the effect of the automatic surveillance system on surveillance rate of patients with gastric premalignant lesions. The enrolled patients were divided into group A with intelligent surveillance system, group B with manual reminder, and group C with natural state. The surveillance among the three groups will be compared.

The adherence of doctors to published surveillance guidelines for patients with gastric premalignant lesions varies greatly, and surveillance of patients is critical but time-consuming. In previous studies we developed an automatic surveillance (AS) system to accurately identify patients with gastric premalignant lesions, assign surveillance intervals for different risks of patients and proactively follow up with patients in time. In this study, we proposed a prospective study about the effect of the automatic surveillance system on surveillance rate of patients with gastric premalignant lesions. The enrolled patients were divided into group A with intelligent surveillance system, group B with manual reminder, and group C with natural state. The surveillance among the three groups will be compared.

Inclusion Criteria:

• Male or female aged 18 years or older who undergo upper endoscopy.

Exclusion Criteria:

• 1)No contact information or invalid contact information.
• 2) The surveillance interval cannot be determined according to the surveillance guidelines, including no upper gastrointestinal pathology, therapeutic endoscopy or with history of previous gastrectomy, esophagectomy, or ESD, no dysplasia degrees, no biopsy sites, non-epithelial lesions, duodenal lesions, ulcer and so on.
• 3) Needless for surveillance or others.
• 4) High-grade intraepithelial neoplasia or cancer of the esophagus or stomach.
• 5) Low-grade intraepithelial neoplasia of the esophagus and Barrett's esophagus.
• 6) High-risk diseases or other special conditions for which the patient is deemed unsuitable for clinical trials by the investigator.