The neurodegenerative progression of Parkinson’s disease affects brain structure and function and, concomitantly, alters the topological properties of brain networks. The network alteration accompanied by motor impairment and the duration of the disease has not yet been clearly demonstrated in the disease progression. In this study, we aim to resolve this problem with a modeling approach using the reduced Jansen-Rit model applied to large-scale brain networks derived from cross-sectional MRI data. Optimizing whole-brain simulation models allows us to discover brain networks showing unexplored relationships with clinical variables. We observe that the simulated brain networks exhibit significant differences between healthy controls ( n = 51) and patients with Parkinson’s disease ( n = 60) and strongly correlate with disease severity and disease duration of the patients. Moreover, the modeling results outperform the empirical brain networks in these clinical measures. Consequently, this study demonstrates that utilizing the simulated brain networks provides an enhanced view of network alterations in the progression of motor impairment and identifies potential biomarkers for clinical indices. Author Summary Understanding the progression of neurodegenerative diseases is of extreme importance in medicine. We utilize biophysical whole-brain models to describe how the brain networks change in Parkinson’s disease (PD). We demonstrate clear correlations between the severity of motor impairment and the properties of the simulated brain networks, which are not prominent in empirical brain networks. Furthermore, we show that healthy participants exhibit a pronounced adaptation of network efficiencies in response to varying parameters of the model, while such an adaptation process is suppressed in PD patients with higher disease severity and duration. Our findings suggest a potential model-based biomarker for classification and clinical evaluation of progressive PD using cross-sectional clinical MRI data.

Author Summary In recent years, new insights into brain network communication related to cognitive performance differences in older age have been gained. Simultaneously, an increasing number of studies has turned to machine learning (ML) approaches for the development of biomarkers in health and disease. Given the increasing aging population and the impact cognition has on the quality of life of older adults, automated markers for cognitive aging gain importance. This study addressed the classification and prediction power of resting-state functional connectivity (RSFC) strength measures for cognitive performance in healthy older adults using a battery of standard ML approaches. Classifiability and predictability of cognitive abilities was found to be low across analytic choices. Results emphasize limited potential of these metrics as sole biomarker for cognitive aging. Abstract Age-related cognitive decline varies greatly in healthy older adults, which may partly be explained by differences in the functional architecture of brain networks. Resting-state functional connectivity (RSFC) derived network parameters as widely used markers describing this architecture have even been successfully used to support diagnosis of neurodegenerative diseases. The current study aimed at examining whether these parameters may also be useful in classifying and predicting cognitive performance differences in the normally aging brain by using machine learning (ML). Classifiability and predictability of global and domain-specific cognitive performance differences from nodal and network-level RSFC strength measures were examined in healthy older adults from the 1000BRAINS study (age range: 55–85 years). ML performance was systematically evaluated across different analytic choices in a robust cross-validation scheme. Across these analyses, classification performance did not exceed 60% accuracy for global and domain-specific cognition. Prediction performance was equally low with high mean absolute errors ( MAE s ≥ 0.75) and low to none explained variance ( R 2 ≤ 0.07) for different cognitive targets, feature sets, and pipeline configurations. Current results highlight limited potential of functional network parameters to serve as sole biomarker for cognitive aging and emphasize that predicting cognition from functional network patterns may be challenging.