Combined Intrinsic Local Functional Connectivity With Multivariate Pattern Analysis to Identify Depressed Essential...

Area of Science:

• Neuroscience
• Psychiatry
• Medical Imaging

Background:

• Depression is a common neuropsychiatric symptom in essential tremor (ET).
• Biomarkers and intrinsic brain activity for depressed ET remain unclear.
• Essential tremor (ET) affects motor function and can co-occur with depression.

Purpose of the Study:

• To identify depressed ET patients using multivariate pattern analysis (MVPA) and local brain functional connectivity.
• To explore intrinsic brain activity differences in depressed ET.
• To investigate potential diagnostic biomarkers for depressed ET.

Main Methods:

• Utilized voxel-level local brain functional connectivity (regional homogeneity, ReHo) mapping.
• Applied binary support vector machine (BSVM) and multiclass Gaussian Process Classification (MGPC) algorithms.
• Classified 41 depressed ET, 43 non-depressed ET, and 45 healthy controls (HCs).

Main Results:

• MGPC achieved 84.5% accuracy for classifying three groups.
• BSVM showed high accuracy (up to 90.7%) in distinguishing ET groups from HCs and between ET groups.
• Identified distinct brain pathways (cerebellar-motor-prefrontal gyrus-anterior cingulate cortex) and correlations between ReHo values and depression severity.

Conclusions:

• Combined ReHo maps with MVPA can identify depressed ET.
• This method reveals intrinsic brain activity changes in depressed ET.
• Findings suggest a potential diagnostic biomarker for depressed ET.