Temporal-spatial dynamic functional connectivity analysis in schizophrenia classification.

Area of Science:

• Neuroscience
• Medical Imaging
• Computational Psychiatry

Background:

• Resting-state functional magnetic resonance imaging (rs-fMRI) measures brain activity.
• Functional connectivity networks (FCNs) show promise for identifying neuropsychiatric disorders like schizophrenia (SZ).
• Traditional static FCNs overlook dynamic connectivity changes and complex interactions.

Purpose of the Study:

• To develop and evaluate dynamic functional connectivity networks (DFCNs) and high-order functional connectivity networks (HFCNs).
• To improve the identification and classification of schizophrenia using advanced FCN analysis.
• To explore potential biomarkers for SZ based on aberrant brain connectivity.

Main Methods:

• Constructed DFCNs to capture temporal variations in brain connectivity.
• Developed HFCNs based on DFCNs to analyze complex, multi-region interactions.
• Extracted temporal variability and high-order network topology features from DFCN and HFCN.

Main Results:

• The proposed DFCN and HFCN methods achieved a classification accuracy of 81.82% for SZ identification.
• This accuracy significantly outperformed competing methods.
• Feature analysis identified potential biomarkers related to aberrant connectivity in SZ.

Conclusions:

• Dynamic and high-order FCN analysis offers a more effective approach for SZ identification than static methods.
• The extracted features from DFCN and HFCN can serve as potential biomarkers for SZ.
• This methodology advances the use of rs-fMRI in diagnosing and understanding schizophrenia.