A Prospective Real-World Study of Pathology Artificial Intelligence for Predicting Molecular Alterations in Gliomas
Trial Snapshot
- Phase
- Not Applicable
- Status
- Recruiting
- Enrollment
- 2,000
- Locations
- 1
- Primary Endpoint
- Accuracy of AI model in predicting key molecular alterations in glioma
Study Overview
Brief Summary
The goal of this clinical study is to learn if an artificial intelligence (AI) model can accurately predict important molecular changes in gliomas, a type of brain tumor, using digital pathology images.
The main questions this study aims to answer are:
How accurate is the AI model in predicting key molecular alterations compared with standard molecular testing? Can the AI model shorten the time needed for diagnosis and reduce the need for expensive molecular tests?
Researchers will collect whole slide images from multiple hospitals and use the AI model to predict molecular results. The predictions will be compared with the actual test results from standard laboratory methods.
Participants will:
Allow the use of their pathology images and molecular test results for research.
Have no additional treatments or procedures beyond standard medical care.
This study will help determine whether AI-assisted tools can provide faster and lower-cost molecular diagnosis for glioma, improving patient care and supporting equal access to precision medicine.
Study Design
- Study Type
- Observational
- Observational Model
- Cohort
- Time Perspective
- Prospective
Eligibility Criteria
- Ages
- 18 Years to 100 Years (Adult, Older Adult)
- Sex
- All
- Accepts Healthy Volunteers
- No
Inclusion Criteria
- •Participant (or legally authorized representative) has voluntarily signed the informed consent form.
- •Age ≥ 18 years at the time of enrollment.
- •Histologically suspected diffuse glioma based on biopsy or surgical resection.
- •Availability of complete clinical information and usable digital pathology slides with hematoxylin and eosin (H&E) staining.
- •Postoperative molecular pathology results available for comparison.
Exclusion Criteria
- •Poor-quality pathology samples (e.g., insufficient tissue, large folding or contamination of slides, or substandard digital scanning quality).
- •Determined by the investigator to be unsuitable for participation in the study for any reason.
Outcomes
Primary Outcomes
Accuracy of AI model in predicting key molecular alterations in glioma
Time Frame: Within 1 week after whole slide images (WSIs) are obtained
The primary outcome is the diagnostic performance of the AI-based pathology model in predicting key molecular alterations in glioma. Accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) will be calculated by comparing AI predictions with reference results from standard molecular pathology testing.
Secondary Outcomes
No secondary outcomes reported