Structural Connectivity Alterations Within Frontal Subregions in Parkinson's Disease: Implications for Motor Dysfunction.
Authors
Affiliations (5)
Affiliations (5)
- College of Control Science and Engineering, China University of Petroleum (East China), Qingdao 266580, PR China (Y.Y., W.L., Y.W.).
- Institute for Digital Medicine and Computer-assisted Surgery, Qingdao University, Qingdao, PR China (X.C.); Shandong Provincial Key Laboratory of Digital Medicine and Computer-assisted Surgery, The Affiliated Hospital of Qingdao University, Qingdao, PR China (X.C.).
- Department of Radiology, The Affiliated Hospital of Qingdao University, Qingdao, PR China (Y.R., M.L.).
- College of Control Science and Engineering, China University of Petroleum (East China), Qingdao 266580, PR China (Y.Y., W.L., Y.W.). Electronic address: [email protected].
- School of Medicine, Queen's Medical Centre, University of Nottingham, Nottingham, UK (M.K.); NIHR Biomedical Research Centre, University of Nottingham, Nottingham, UK (M.K.).
Abstract
Parkinson's disease (PD) is characterized by disrupted basal ganglia-thalamo-cortical connectivity, yet how frontal network topology relates to motor phenotype heterogeneity remains unclear. Conventional atlas-based approaches average out microscale organization within heterogeneous cortical areas. This study aimed to characterize fine-grained frontal subregional connectivity alterations in PD using high-resolution structural network analysis. Twenty-three patients with PD and 22 age- and sex-matched controls underwent diffusion tensor imaging and T1-weighted magnetic resonance imaging. Individual high-resolution frontal networks were constructed, and graph-theoretical metrics were computed for each frontal subregion to quantify connectivity patterns. Consensus Louvain clustering assessed modularity, and machine learning with SHapley Additive exPlanations interpretation performed individual classification. Patients with PD exhibited heterogeneous topological alterations across multiple frontal subregions (corrected p < 0.05). The left rostral middle frontal gyrus (RMF.L) showed increased density and degree, whereas the lateral and medial orbitofrontal cortices showed decreases. RMF.L further exhibited a greater number of modules together with an altered modular organization. These topological alterations were significantly correlated with UPDRS-III motor scores. Topological features achieved a diagnostic accuracy of 91.69% for PD diagnosis and 91.90% for subtype differentiation, exceeding the performance of connectivity-based features. Intraregional topological reorganization within the frontal lobe, most prominently in RMF.L, was associated with motor impairment in PD and may reflect maladaptive network remodeling. High-resolution network analysis may provide a sensitive framework for characterizing intraregional structural reorganization in PD and warrants further validation as an adjunctive imaging marker for disease stratification.