Functional Connectivity Changes in Multiple-Frequency Bands in Acute Basal Ganglia Ischemic Stroke Patients: A...

Area of Science:

• Neuroimaging
• Stroke Research
• Brain Connectivity

Background:

• Resting-state functional connectivity (rs-FC) studies have explored changes in the primary motor cortex (M1) in acute basal ganglia ischemic stroke (BGIS).
• However, a clear understanding of frequency-specific rs-FC alterations in M1 for BGIS patients remains elusive.

Purpose of the Study:

• To investigate altered M1 rs-FC in three distinct frequency bands (conventional, slow-5, slow-4) in acute BGIS patients.
• To evaluate the potential of these frequency-specific rs-FC changes as biomarkers for BGIS identification using a support vector machine (SVM).

Main Methods:

• Functional magnetic resonance imaging (fMRI) was used to assess rs-FC in 28 acute BGIS patients and 42 healthy controls (HCs).
• Seed-based analysis focused on bilateral M1 regions of interest (ROIs).
• Voxel-wise FC values across conventional, slow-5, and slow-4 frequency bands were extracted as features for SVM classification.

Main Results:

• Significant rs-FC alterations were observed in both ipsilesional and contralesional M1 in BGIS patients.
• Specific changes included decreased FC with the lingual gyrus and increased FC with the medial superior frontal gyrus (SFGmed) in various frequency bands.
• The SVM model, utilizing combined slow-4 and slow-5 frequency band features, achieved the highest classification accuracy with an AUC of 0.86.

Conclusions:

• Acute BGIS is associated with frequency-specific alterations in M1 rs-FC.
• SVM analysis demonstrates promise in identifying and classifying these frequency-dependent connectivity changes.
• Identified abnormal brain regions may serve as therapeutic targets for stroke rehabilitation.