Chaos theory based mathematical modelling as manifested from scalp EEG using frequency analysis

The EEG (Electroencephalogram) signals are brain mapped signals that contain information about the brain´s complexity and uncertainty. The EEG signals though are useful but due to a large variety of data in them, may look random in nature. We have to extract the proper information from the data by computational modelling of scalp EEG signals. The chaos theory helps in analysing the neurobiological parameters which include Lyapunov Exponent, Approximate Entropy and Hurst Exponent. The frequency filtering of the data helps us in calculation of parameters for different frequency range. It is found that the different classes of data can be catalogued by computing parameters in a specific range of frequencies. As the different frequency band represents different states of mind so the value of parameter for subjects of same classes exhibit same pattern and can be easily distinguished from the other class of benign subjects. Moreover the data is testified by the values of Hurst Exponent for auto correlation.