Functional Connectivity Evaluation in Run Time Using Graph Theory
Graph theory has been widely used to provide information and represent the relationship between objects or elements in different contexts. Recently, ways of transforming time series into graphs have attracted the attention of researchers, mainly intrigued by the possibility of characterizing processes (eg physiological states) through their classical measurements (degree, eigenvector centrality, centrality, etc.). In this sense, this master's thesis aims to graphically represent the functional connectivity in execution time through graphs, in order to analyze EEG patterns collected in the context of brain-computer interface (BCI) systems through the dynamics of graph-based metrics. To achieve these goals, a web service was developed on Node.js technology that captures the EEG signals coming from the OpenBCI headset and makes them available in a network. An interface has also been created in the JAVA language and using the GraphStream tool, which captures these signals and transforms them into graphs with several configuration options that are updated at runtime.