Vision-language AI to diagnose brain diseases from imaging and clinical data
A Vision-Language Foundation Model for Brain Disease Diagnosis From Multimodal Data
This project will test a new AI that reads scans, pathology, genomics, and clinical records to help diagnose a wide range of brain diseases for patients with tumors, vascular problems, or neurodegeneration.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 100000 (estimated) |
| Sex | All |
| Sponsor | Xiangya Hospital of Central South University Academic / other |
| Locations | 1 site (Changsha, Hunan) |
| Trial ID | NCT07126821 on ClinicalTrials.gov |
What this trial studies
This observational project will compile patient demographics, multimodal medical imaging (MRI, CT, PET), histopathology, genomic data when available, and other laboratory tests to train a vision-language foundation model. The model is being built to be explainable, provide uncertainty estimates, and automatically generate diagnostic-style reports to support clinicians. The dataset will include pathologically confirmed brain tumor cases, clinically confirmed other brain diseases, and non-brain-disease controls with complete imaging. An exploratory aim is to validate model performance using large-scale MRI cohorts for generalizability.
Who should consider this trial
Good fit: Ideal candidates are patients with accurately diagnosed brain diseases and complete clinical and imaging records, as well as control patients who have complete brain MRIs and no history of brain disease.
Not a fit: Patients with incomplete or heavily artifacted imaging, missing key clinical data, or cases outside the represented disease spectrum are unlikely to benefit and may be excluded.
Why it matters
Potential benefit: If successful, the AI could help clinicians make faster, more consistent, and more transparent diagnoses across many brain diseases and produce draft reports to streamline care.
How similar studies have performed: AI tools have shown promising results for specific imaging tasks such as tumor segmentation and stroke detection, but a broad vision-language foundation model covering all brain diseases is largely novel and unproven.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: Patients with brain diseases: * Patients with brain tumors were pathologically diagnosed. * Patients with other brain diseases were correctly diagnosed. * The clinical case data of all patients were complete. Non-brain disease population: * All patients have complete clinical case data, complete brain MRI, no history brain diseases, no brain surgery or other brain diseases that affect the diagnosis and observation of MR imaging. Exclusion Criteria: * Cases in which MRI were incomplete or with significant noise and artifacts.
Where this trial is running
Changsha, Hunan
- Xiangya Hospital of Central South University — Changsha, Hunan, China (Recruiting)
Study contacts
- Study coordinator: Xuan Gong, PhD.
- Email: gong.xuan@csu.edu.cn
- Phone: 0086-731-8975-3037
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.