abstract
- © 2016, Springer-Verlag France.Electro EncephaloGraphic (EEG) signals follow complex patterns, which are needed to analyse either for healthy or sick people. Instead of analyzing long rolls of EEG, to decide how a brain works given a task to solve, a index which can simplify massive data (EEG) information in a very simple plot in terms of an index: Average Bivariate MultiScale Entropy (ABMSE) This index is a modified version of a nonlinear statistic known as MultiScale Entropy (MSE) that resulted to be very useful for this task is exploited. Early results show the proposal statistic ABMSE concentrates massive complexity information in a single quantity that can be plotted per brain zone. Some ideas are discussed on the application of interactive engineering and design. In particular, the use of learning algorithms to distinguish different pathological scenarios based on EEG signal post processing. This will support neurologists, bioengineers and neuroscientists in healthy or sick people.