Automatic processing protocol to evaluate the impact of functional network damage and reorganization on cognitive functions after stroke

An ischemic stroke is a local lesion that disrupts the large-scale structural and functional connectivity of the brain. Although local, the ischemic stroke often leads to deficits in cognitive functions which can’t be explained by local brain damage. It is believed that stroke-induced large-scale network alteration represents the mechanisms responsible for a decline in cognitive functions which are dependent on large-scale integration. To gain insight into the pathophysiological principles of how a local lesion results in a global cognitive decline requires a reliable and robust algorithm that can quantify the relationship between cognitive functions and network properties. In this study, we have developed, optimized, and tested a processing pipeline to parameterize complex neuropsychological evaluation and determine the functional connectivity from high-density EEG recordings. The developed algorithm was applied on a cohort of 27 patients who suffered a stroke and who were underwent cognitive examinations and high-density EEG monitoring one and two years after the stroke. The developed automatic algorithm demonstrated that it can reliably estimate functional connectivity and that it is robust against the physiological and technical artifacts. The proposed processing pipeline allows an unbiased and quantitative characterization of cognitive performance and its comparison with functional connectivity alterations.