Assessment of Directional Connectivity Between Neural Sources Using Effective Connectivity Measures and Particle Filters

Abstract

Electrical neural activity monitoring and recording will increase our understanding of how the human brain works. Tracking mechanisms of neural activity led to better diagnosis and management of severe neurological conditions such as Parkinson’s disease and epilepsy. More importantly, these approaches were used to distinguish between various types of seizures based on the location and direction of the seizure foci, thereby increasing the outcomes of epilepsy surgery. A detailed study was carried out on the role of neural synchrony in brain functions with Electroencephalography (EEG). Most studies had been conducted on EEG connectivity analysis at sensor level. It is not easy to evaluate the connected networks because the volume conductive effect significantly distorts signals because of the electrical conductiveness of the head and often scalp electrodes derive input from the same sources in the brain. These factors help to estimate the real connectivity between brain regions inaccurately. The suggested approach is referred to as EEG source connectivity. The inverse problem is the estimation of the localized current dipole model from the EEG measurements. In order to solve the inverse EEG problem, advanced signal-processing algorithms such as the efficient implementation of PF have been built to facilitate direct exposure to neural dipole sources in real time and measure the connectivity of neural sources time courses using functional and effective connectivity measures.