INTRODUCTION: Brain network integration has been associated with higher-order functioning. Changes in connectome features, such as decrease in global integration, are linked to successful memory (Davis et al., 2018), as well as social, health, and cognitive factors (Powell et al., 2018). This association has also been found in Parkinson's Disease (PD) patients (Baggio et al. 2014). Understanding whether cognitive function is reflected in network efficiency measures, could serve as a potential biomarker for neurodegenerative diseases.

AIM(S): This analysis aims to determine whether network integration metrics will characterize healthy functioning individuals in comparison to individuals with PD, and examine whether cognitive performance is correlated with functional brain network integration.

METHOD(S): The dataset includes three groups of subjects (age (60- 94; M=69; SD=7.4)): PD with normal cognition (PD-NC, n = 18), PD with Mild Cognitive Impairment (PD-MCI, n= 15), and Healthy Controls (HC, n = 22). For each subject relevant cognitive performance measurements were collected during a 2-year longitudinal study. From the resting-state recordings functional brain networks were constructed based on ROI-to-ROI averaged connectivity and measures of network integration were extracted at a cost threshold of 0.15.

RESULTS: Comparisons between HC and PD-NC, as well as between PD-NC and PD-MCI, did not show significant differences. One comparison indicated reduced global efficiency in the PD-MCI group compared to HC (U = 239, ? < 0.05). In addition, female subjects had higher global efficiency (U = 188, ? < 0.005), which may be due to imbalanced representation in the PD-MCI group.

CONCLUSIONS: Lack of differences between the groups suggests that in chosen metrics suggests that global network integration does not capture the effects related to Parkinson's Disease at a given threshold, and that potentially local network integration metrics, as well as connections between specific regions of interests should be investigated instead.