Resting-State Functional Magnetic Resonance Imaging for Language Preoperative Planning

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Clinical Neurology

Background:

• Functional magnetic resonance imaging (fMRI) is crucial for preoperative language mapping in neurosurgery.
• Task-based fMRI faces limitations like poor performance and motion artifacts in clinical settings.
• Resting-state fMRI protocols offer a potential alternative for pre-surgical language mapping.

Purpose of the Study:

• To evaluate the feasibility of using resting-state fMRI for language network mapping.
• To compare resting-state fMRI language maps with conventional task-based fMRI maps.
• To assess the accuracy of an automated classification method for resting-state language networks.

Main Methods:

• Fifteen patients with brain lesions underwent both task-based and resting-state fMRI.
• Task-related activity was analyzed using general linear model and independent component analysis (ICA).
• Resting-state networks were extracted via ICA and classified manually and automatically.

Main Results:

• Automated classification of resting-state networks accurately identified language networks, matching expert manual classification.
• A significant overlap was observed between task-related activity and resting-state language maps, especially in language regions of interest.
• Resting-state fMRI maps demonstrated comparable sensitivity and higher specificity than task-related maps.

Conclusions:

• Resting-state fMRI protocols are a viable and efficient alternative for pre-surgical language mapping.
• Automated classification enhances the clinical utility of resting-state fMRI for language network identification.
• This approach may overcome limitations associated with task-based fMRI in neurosurgical patients.