New Graph-Theoretical-Multimodal Approach Using Temporal and Structural Correlations Reveals Disruption in the...

Area of Science:

• Neuroscience
• Psychiatry
• Network Science

Background:

• Schizophrenia is characterized by widespread alterations in brain network connectivity.
• Current models often analyze functional and structural connectivity independently, potentially missing integrated network dynamics.
• Combining functional and structural information may yield more comprehensive brain network models.

Purpose of the Study:

• To introduce and validate a novel graph-theoretical multimodal model (FD) for constructing brain networks.
• To compare the efficacy of the FD model against traditional pure functional connectivity (pFC) analysis in detecting schizophrenia-related network alterations.
• To investigate disrupted connectivity patterns in patients with schizophrenia using integrated neuroimaging data.

Main Methods:

• Applied a flexible graph-theoretical multimodal model (FD) integrating functional connectivity (F) and structural connectivity (D) matrices.
• Utilized magnetic resonance imaging (MRI) data, including structural and resting-state scans, from schizophrenia patients (n=35) and healthy controls (n=41).
• Performed a comparative analysis using traditional pure functional connectivity (pFC) as a reference.

Main Results:

• The FD model revealed significant disruptions in thalamo-cortical network connectivity in schizophrenia patients.
• The pFC analysis failed to detect significant group differences after multiple comparison correction.
• The FD model demonstrated superiority in highlighting relevant functional, structural, and topological network features.

Conclusions:

• The FD model offers a more sensitive and reliable approach to analyzing brain network alterations in schizophrenia compared to pFC.
• This multimodal approach can capture subtle connectivity changes underlying complex psychiatric disorders.
• The FD model has broad applicability beyond resting-state fMRI, including EEG and MEG data.