Evidence of self-organization in brain electrical activity using wavelet-based informational tools

In the present work, we show that appropriate information-theory tools based on the wavelet transform (relative wavelet energy; normalized total wavelet entropy, H; generalized wavelet complexity, CW), when applied to tonic-clonic epileptic EEG data, provide one with valuable insights into the dynamics of neural activity. Twenty tonic-clonic secondary generalized epileptic records pertaining to eight patients have been analyzed. If the electromyographic activity is excluded the difference between the ictal and pre-ictal mean entropic values (ΔH=〈H(ictal)〉-〈H(pre-ictal)〉) is negative in 95% of the cases (p<0.0001), and the mean complexity variation (ΔCW=〈CW(ictal)〉-〈CW(pre-ictal)〉) is positive in 85% of the cases (p=0.0002). Thus during the seizure entropy diminishes while complexity grows. This is construed as evidence supporting the conjecture that an epileptic focus in this kind of seizures triggers a self-organized brain state characterized by both order and maximal complexity.