AI-Driven Multimodal Imaging Integration for Diagnosis and Prognostication of Digestive System Diseases

First Affiliated Hospital, Sun Yat-Sen University5,000 enrolled

Overview

The goal of this observational study is to to develop a noninvasive disease assessment system by leveraging artificial intelligence (AI) to comprehensively analyze multi-modal imaging features, including magnetic resonance enterography (MRE) and computed tomography enterography (CTE), for the diagnosis and prognostication of digestive diseases. Participants will be randomly assigned to either conventional endoscopy or virtual endoscopy groups. The predictive performance of both groups for prognostic indicators, such as clinical remission rate and recurrence risk, will be compared during follow-up to verify the non-inferiority of the virtual endoscopy group.

Study Type

OBSERVATIONAL

Enrollment

5,000

Conditions

Digestive Diseases Radiology AI (Artificial Intelligence)Imaging

Interventions

The non-inferiority of the virtual endoscopy group will be verified by comparing the predictive performance for prognostic indicators (e.g., clinical remission rate, recurrence risk) between the two groups during follow-up.

Eligibility

Sex: ALLHealthy volunteers:

Medical Language ↔ Plain English

Inclusion Criteria: * Patients with multimodal-confirmed diagnoses (clinical, imaging, endoscopic, and pathological) of: * Inflammatory bowel disease (IBD; Crohn's disease or ulcerative colitis) * Intestinal tuberculosis * Behçet's disease * Availability of ≥1 technically adequate CT or MR scan with high-quality colonoscopy performed within ±1 month of imaging. Exclusion Criteria: * ・Suboptimal imaging quality (e.g., low-dose artifacts, metal artifacts) * Inadequate bowel preparation for endoscopy * Incomplete examinations due to poor tolerance

Outcomes

Primary Outcomes

The area under the ROC curve (AUC) to assess the performance of diagnostic model.

After baseline MR or CT scanning, patients were followed up.

Time frame: 6 months