Brain connectivity for subtypes of parkinson's disease using structural MRI

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Neurology

Background:

• Parkinson's disease (PD) classification typically relies on clinical symptoms, potentially overlooking underlying neurobiological heterogeneity.
• An imaging-based, data-driven approach offers a novel method for subtyping PD patients.

Purpose of the Study:

• To subtype Parkinson's disease patients using grey matter information from structural MRI.
• To perform comparative structural brain connectivity analysis between identified PD subtypes.

Main Methods:

• Source-based morphometry applied to MRI scans of 131 PD patients and 78 controls.
• Subtyping based on component loadings, followed by construction of subtype-specific structural brain connectivity matrices.
• Network metrics analysis (clustering coefficient, efficiency, centrality) using graph theory.

Main Results:

• Identified two distinct PD subtypes (A and B) and a common subtype (AB) based on grey matter distribution.
• Subtype A showed frontal lobe weighting; Subtype B showed temporal lobe weighting.
• Significant inter-subtype differences in network metrics (clustering coefficient, efficiency, participation coefficient, betweenness centrality) were observed.

Conclusions:

• MRI-based data-driven subtyping highlights the critical roles of frontal and temporal lobes in PD.
• Graph theory analysis reveals differential structural brain connectivity architectures across PD subtypes.
• These findings suggest potential for developing subtype-specific therapeutic strategies for Parkinson's disease.