Skip to content

Research on the Construction of a Multi-modal Fusion Diagnostic Model for Precancerous Lesions of Gastric Cancer Based on Tongue Image AI and Liquid Biopsy

Research on the Construction of a Multi-modal Fusion Diagnostic Model for Precancerous Lesions of Gastric Cancer Based on Tongue Image AI and Liquid Biopsy

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ITMCTR
Registry ID
ITMCTR2026001222
Enrollment
Unknown
Registered
2026-05-14
Start date
2026-06-01
Completion date
Unknown
Last updated
2026-06-01

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

Conditions

Precancerous lesion of gastric cancer

Interventions

Lesion group:Not applicable

Sponsors

Shenzhen Traditional Chinese Medicine Hospital
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1) Participants aged 18 years or older were all subjected to gastroscopy. Based on endoscopic and pathological examinations, they were diagnosed as having normal gastroscopy (control group) or atrophic gastritis combined with intestinal metaplasia (lesion group). 2) They voluntarily participated in the study and signed the informed consent form. 3) Diagnostic criteria: The control group consisted of those who were diagnosed with chronic superficial gastritis or no obvious lesions through endoscopic examination. The lesion group included those who had gastric mucosal lesions detected by endoscopy and underwent gastric mucosal biopsy, and whose pathological diagnosis was precancerous lesions (the focus of this study was on atrophy combined with intestinal metaplasia). The pathological diagnosis was independently reviewed by two pathologists, and in case of inconsistent diagnoses, a third pathologist made the final determination. The pathologists were unaware of the diagnosis results and serum test results of the study participants during the diagnosis process.

Exclusion criteria

Exclusion criteria: 1) Patients with confirmed gastric cancer; 2) Those whose tongue lesions affect the collection of tongue images; 3) Patients with severe liver or kidney dysfunction; 4) Those who have recently taken drugs that affect gastric function (such as proton pump inhibitors, antibiotics, etc.); 5) Patients with severe systemic diseases or acute or chronic infections.

Design outcomes

Primary

MeasureTime frame
The model's discrimination ability ;The model's classification performance;

Secondary

MeasureTime frame
Clinical consistency evaluation of the constructed model;Clalibration degree and calibration curve of the constructed model;the results of difference analysis of the constructed model;

Countries

China

Contacts

Public ContactJing-wei Li

Shenzhen Traditional Chinese Medicine Hospital

lijingwei120@126.com13430328517

Outcome results

None listed

Source: ITMCTR (via WHO ICTRP) · Data processed: Jun 11, 2026