2019/12/05 by Tiziana Cattai, Cattai, Tiziana, Stefania Colonnese +9 · 1 citation
Neuroscience · Engineering · #EEG and Brain-Computer Interfaces #Functional Brain Connectivity Studies #Advanced Memory and Neural Computing
paper · pdf · doi:10.48550/arxiv.1912.02745
The extraction of brain functioning features is a crucial step in the\ndefinition of brain-computer interfaces (BCIs). In the last decade, functional\nconnectivity (FC) estimators have been increasingly explored based on their\nability to capture synchronization between multivariate brain signals. However,\nthe underlying neurophysiological mechanisms and the extent to which they can\nimprove performance in BCI-related tasks, is still poorly understood. To\naddress this gap in knowledge, we considered a group of 20 healthy subjects\nduring an EEG-based hand motor imagery (MI) task. We studied two\nwell-established FC estimators, i.e. spectral- and imaginary-coherence, and\ninvestigated how they were modulated by the MI task. We characterized the\nresulting FC networks by extracting the strength of connectivity of each EEG\nsensor and compared the discriminant power with respect to standard power\nspectrum features. At the group level, results showed that while\nspectral-coherence based network features were increasing the controlateral\nmotor area, those based on imaginary-coherence were decreasing. We demonstrated\nthat this opposite, but complementary, behavior was respectively determined by\nthe increase in amplitude and phase synchronization between the brain signals.\nAt the individual level, we proved that including these network connectivity\nfeatures in the classification of MI mental states led to an overall\nimprovement in accuracy. Taken together, our results provide fresh insights\ninto the oscillatory mechanisms subserving brain network changes during MI and\noffer new perspectives to improve BCI performance.\n