Functional connectivity-based searchlight multivariate pattern analysis for discriminating Parkinson's disease...

Area of Science:

• Neuroscience
• Medical Imaging
• Computational Biology

Background:

• Parkinson's disease (PD) and PD with mild cognitive impairment (MCI) are characterized by disruptions in brain functional connectivity (FC).
• Differentiating these conditions and understanding their clinical severity is crucial for research and treatment.

Purpose of the Study:

• To differentiate Parkinson's disease (PD) patients from healthy controls (HCs).
• To distinguish PD patients with mild cognitive impairment (PD-MCI) from those with normal cognition (PD-NC).
• To evaluate the predictive capability of connectivity-based searchlight multivariate pattern analysis (CBS-MVPA) for clinical severity in PD.

Main Methods:

• Utilized resting-state functional magnetic resonance imaging (fMRI) data from 261 participants in the Parkinson's Progression Markers Initiative (PPMI).
• Applied CBS-MVPA to examine whole-brain FC subnetworks for group discrimination and prediction of clinical measures.
• Included 98 PD-MCI patients, 98 PD-NC patients, and 65 HCs in the analysis.

Main Results:

• CBS-MVPA identified 20 FC subnetworks distinguishing PD from HCs with moderate accuracy (71.25%–75.48%).
• PD-MCI was distinguished from PD-NC with modest accuracy (57.21%–62.74%).
• Specific subnetworks showed modest predictive value for motor severity and associations with cognitive performance, though accuracy was near chance level.

Conclusions:

• Introduced a novel CBS-MVPA approach for identifying distributed FC patterns in PD and PD-MCI.
• Detected subnetworks contain measurable predictive signals for diagnostic status and clinical severity.
• Highlights the potential of whole-brain multivariate FC analyses for PD research and therapeutic interventions.