A potential diagnostic biomarker for schizophrenia based on local functional connectivity using dynamic regional...

Area of Science:

• Neuroimaging
• Psychiatry
• Computational Neuroscience

Background:

• Schizophrenia (SZ) diagnosis faces challenges due to symptomatic overlap with other mental disorders like bipolar disorder (BD).
• Objective neuroimaging markers are needed for accurate SZ diagnosis.
• Current diagnostic criteria for SZ can lead to inaccuracies.

Purpose of the Study:

• To investigate dynamic regional phase synchrony (DRePS) as a potential neuroimaging biomarker for SZ.
• To differentiate SZ from healthy controls (HC) and BD using DRePS.
• To assess the correlation of DRePS with clinical symptoms in SZ.

Main Methods:

• Functional magnetic resonance imaging (fMRI) and T1 data were collected from 931 participants (300 SZ, 145 BD, 486 HC).
• Dynamic regional phase synchrony (DRePS) of BOLD signals was analyzed.
• Support vector machine (SVM) models were trained and tested for classification tasks.

Main Results:

• SZ patients exhibited significantly reduced DRePS compared to HC.
• Significant differences in DRePS were observed between SZ and BD patients.
• DRePS showed a significant correlation with clinical symptom scores (PANSS) in SZ.
• SVM models achieved high accuracy in classifying SZ from HC (up to 85%) and SZ from BD (approx. 72%).

Conclusions:

• DRePS of BOLD signals is a promising neuroimaging biomarker for differentiating SZ from both HC and BD.
• DRePS is correlated with clinical symptoms, suggesting its prognostic value.
• This finding could improve diagnostic accuracy and clinical management of SZ.