A Novel Machine Learning Model for the Detection of Epilepsy and Epileptic Seizures Using Electroencephalographic Signals Based on Chaos and Fractal Theories

Machine learning is an expanding research area. Its main application in the medical field and particularly detection of epilepsy epileptic seizures through electroencephalographic signals (EEG). It aims to design intelligent framework that enables immediate diagnosis this disease without neurological consultation thus saves lives patients by detecting warning them before it happens. However, as a real-time application, kind faces several challenges such accuracy, fast responses, optimal memory usage. Within context, our work was carried out. We propose new machine based on chaos fractal theories. Two novelties are presented paper. Firstly, we method for signal preprocessing, reconstruct versions studied EEG using derivative determination chaotic injection. Secondly, suggest analysis Higuchi dimension (HFD). In fact, HFDs extracted from derivatives lead detect epilepsy, whereas with injection seizure detection. addition, feature fusion helped linearize all classification problems. An experimental study Bonn database proves efficiency contributions comparison published research. accuracy 100% achieved different cases few features simple linear classifier.