Neurosurgical Neuronavigation Using Resting State MRI and Machine Learning

The Washington University

Status: Enrolling

Conditions: Glioblastoma Multiforme

Treatments: Device: Support Vector Machine

Study type: Observational

Funder types: Other; NIH

Identifiers

NCT05864976

202305045

2R01CA203861 (U.S. NIH Grant/Contract)

Details and patient eligibility

About

This study is investigating the use of a computer algorithm to analyze scans of the brain before surgery to predict how a person's tumor will respond to treatment.

Enrollment: 100 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Must have a radiological diagnosis of a lesion in the brain with characteristics consistent with glioblastoma multiforme.
• Must be planning to undergo a pre-operative MRI.
• Must be at least 18 years old.
• Must be able to understand and willing to sign an IRB approved written informed consent document.

Exclusion criteria

• Contraindication to MRI.
• Inability to have clinical follow-up (e.g., patient is out of town and will do follow-up elsewhere).

Trial design

100 participants in 1 patient group

Standard of care rsfMRI using the Support Vector Machine algorithm

Description:

* Once enrolled, clinical pre-surgical MRI will be done on Siemens 3T Prisma or Skyra scanners using a standard pre-surgical tumor protocol. Resting-state functional MRI (rsfMRI) will be acquired. The Support Vector Machine (SVM) algorithm will be used on this pre-surgical MRI. * Patients will undergo post-operative MRI at approximately 8-12 weeks following surgical resection to evaluate extent of resection. Patients will then undergo subsequent MRI imaging every 2-3 months as part of routine clinical care to monitor for recurrence. The following MR sequences will be acquired: pre-and post-contrast T1-weighted, T2-weighted FLAIR, diffusion weighted imaging. MRI scans will be reviewed by a board-certified neuroradiologist to determine date of radiographic progression/recurrence. Imaging features at recurrence including location, multifocality, and presence of diffuse or distant recurrence will also be recorded.

Treatment:

Device: Support Vector Machine