Multicenter Observational Study of Multimodal AI for Upper GI Mesenchymal Tumor Diagnosis

Sponsor
Huazhong University of Science and Technology
Study ID
NCT07078136
Status
Recruiting
Apply to this trial

Conditions

  • Gastrointestinal Stromal Tumor (GIST)
  • Leiomyoma
  • Schwannoma
  • Submucosal Tumor

Eligibility Criteria

Sex
ALL
Age
18 Years - N/A
Healthy Volunteers
Not accepted

Interventions

  • Multimodal AI Model — DIAGNOSTIC_TEST
    Patients' endoscopic images, EUS images, and clinical data will be analyzed by a multimodal AI model for lesion classification and GIST risk stratification.
  • Expert Endoscopist Assessment — DIAGNOSTIC_TEST
    Endoscopic ultrasound images will be interpreted by experienced endoscopists for comparison with the AI model.

Study Details

This study develops a multimodal AI model using endoscopic ultrasound, white-light endoscopy, and clinical information to support the diagnosis of upper GI mesenchymal tumors and the risk stratification of gastric GISTs.

Key Dates

Start date
Jul 28, 2025
Status verified
Jul 2025
Primary completion
Mar 31, 2026
Completion
Jun 30, 2026

Study Design

Enrollment
130 participants (estimated)

Arms

  • Arm: All Participants
    All enrolled patients with upper gastrointestinal subepithelial lesions confirmed by histopathology. Each participant will undergo standard diagnostic evaluation and independent multimodal AI prediction and expert endoscopist diagnosis.

Primary Outcome Measure

Diagnostic accuracy of a multimodal AI model for differentiating gastrointestinal stromal tumors (GISTs) from other upper gastrointestinal mesenchymal tumors [ Time Frame: After the training process of the multimodal AI model is completed，on average per year ]

Central Contacts

Related Studies