Multicenter Prospective Study on MRI AI Model for Midline Glioma Subtyping and Prognosis:

Trial statusRecruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
AgeNot listed
SponsorXiangya Hospital of Central South University

About this trial

A vision-language model using preoperative MRI and clinical variables has been developed to simultaneously predict three key molecular markers in midline gliomas: H3K27M, IDH, and 1p/19q. This prospective multicenter study will validate the model's accuracy in preoperative molecular subtyping and its value in prognostic assessment and clinical decision-making across multiple neurosurgical centers.

Eligibility criteria

This trial does not accept healthy volunteers

Qualifiers

Patients with diffuse gliomas were pathologically and molecularly diagnosed.

The clinical case data of all patients were complete.

Patients underwent preoperative MRI examination.

Disqualifiers

The tumor is not located in the intracranial midline.

Cases in which MRI were incomplete or with significant noise and artifacts.

Trial population

Patients with diffuse gliomas located in the intracranial midline.

Trial design

Design model

Case-only

Time perspective

Prospective

Treatments tested in this trial

Not listed

Trial groups

No trial groups listed

Trial outcomes

Primary outcomes

1

Diagnostic accuracy for midline glioma molecular subtypes

Model predictions compared with postoperative histopathology and molecular testing (gold standard). Performance metrics include AUC, F1 score, sensitivity, specificity, and accuracy.

Time frame
Perioperative

Secondary outcomes

Other outcomes

Sponsors and contacts

Click on the lead sponsor to view all of their trials.