Stochastic qualifiers of epileptic brain dynamics

Phys Rev Lett. 2007 Mar 30;98(13):138103. doi: 10.1103/PhysRevLett.98.138103. Epub 2007 Mar 28.


We evaluate the capability of reconstructing Fokker-Planck equations for an improved characterization of electroencephalographic (EEG) recordings from epilepsy patients. We derive stochastic qualifiers of brain dynamics that are based on specific characteristics of the Kramers-Moyal coefficients estimated from the EEG. Analyzing long-lasting multichannel EEG recordings from eight patients suffering from focal epilepsies we show that particularly the stochastic part of the dynamics can yield valuable information for diagnostic purposes.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Brain / physiopathology*
  • Electroencephalography / statistics & numerical data*
  • Epilepsies, Partial / physiopathology*
  • Humans
  • Markov Chains
  • Models, Statistical
  • Stochastic Processes