Temporal and spatial variability of dynamic microstate brain-network in early Parkinson's disease

Abstract

The spatiotemporal microstate attributes reflecting motor impairment and cognitive decline exist in early Parkinson’s disease (PD), which are derived from the unknown variation of brain network dynamics underlying each microstate class. To understand the dynamic neural underpinnings of different degree of motor impairment and cognitive decline associated with early PD patients, an improved data-driven brain network analysis based on microstate exploration, was proposed to pinpoint the network dynamical features at local region and global brain levels. The further exploration was to reveal the relation between regional variability and cerebral dysfunction in early PD. We have found the temporal variability of microstate B network in early PD was abnormally increased, and it was mainly distributed in left prefrontal, parietal, and occipital lobe. Comparatively, the temporal variability of microstate C network in early PD was significantly decreased, and it was mainly distributed in frontal and left occipital lobe. In microstate C network, the temporal variability of frontal lobe indicated the cognitive level of early PD patients. In addition, the spatial variability of microstate D network in early PD was significantly decreased, and it was mainly distributed in left prefrontal, bilateral temporal, and left-right parietal lobe. In microstate D network, the abnormal spatial variability of left temporal and right parietal lobe reflected the cognitive and motor impairment of early PD, respectively. Our approach provides innovative methodological support for future research on brain network dynamics, and provides insights into the spatiotemporal interaction patterns of brain activity and their variabilities in early PD.