Multicenter validation of an AI assistant for CT and MRI diagnosis of common systemic diseases
Prospective User Study and Multicenter Validation of Multimodal Medical Imaging Large Models in the Diagnosis of Common Systemic Diseases
This project will test whether an AI imaging tool helps radiologists read CT and MRI scans more accurately and faster for common systemic diseases.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 1000 (estimated) |
| Ages | 18 Years and up |
| Sex | All |
| Sponsor | The Third Affiliated Hospital of Southern Medical University Government |
| Locations | 1 site (Guangzhou, Guangdong) |
| Trial ID | NCT07555002 on ClinicalTrials.gov |
What this trial studies
This is a multicenter, retrospective comparative reader study using large, de-identified CT and MRI datasets drawn from participating institutions. Licensed radiologists will read cases in two sessions: once without AI assistance and once with AI assistance, separated by a washout period to reduce recall bias. Ground truth will be established by expert consensus or pathological results, and outcomes will include AUC, sensitivity, specificity, and per-case reading time. The analysis will compare diagnostic performance and efficiency across radiologists with different experience levels to test robustness across centers.
Who should consider this trial
Good fit: Ideal candidates are patients who underwent CT or MRI for common systemic diseases and whose imaging is available in complete DICOM format with confirmed clinical or pathological reference standards.
Not a fit: Patients with poor-quality or corrupted images, incomplete clinical or pathological reference standards, or very rare conditions not represented in the dataset are unlikely to receive direct benefit from this work.
Why it matters
Potential benefit: If successful, the AI assistant could help radiologists make more accurate and faster diagnoses of common systemic diseases, improving clinical workflow and patient care.
How similar studies have performed: Prior single-center and modality-specific AI reader studies have often shown improved sensitivity and reduced reading time, but large-scale multicenter validation of multimodal large models remains limited.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Patients who underwent systemic medical imaging examinations (e.g., CT or MRI) at participating centers for common systemic diseases. * Imaging data must have confirmed clinical reference standards, expert consensus, or pathological diagnosis. * Availability of complete DICOM format images with standard acquisition protocols. Exclusion Criteria: * Poor image quality (e.g., severe motion or metal artifacts) that precludes definitive diagnosis. * Cases with incomplete clinical or pathological reference standards. * Corrupted image files or duplicate cases.
Where this trial is running
Guangzhou, Guangdong
- The Third Affiliated Hospital of Southern Medical University — Guangzhou, Guangdong, China (Recruiting)
Study contacts
- Principal investigator: Yinghua Zhao, PhD — The Third Affiliated Hospital of Southern Medical University
- Study coordinator: Tao Li, MD
- Email: lt12420131@163.com
- Phone: +86-15527360835
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.