Statistical geometric affinity in human brain electric activity

Phys Rev E Stat Nonlin Soft Matter Phys. 2007 May;75(5 Pt 1):051918. doi: 10.1103/PhysRevE.75.051918. Epub 2007 May 29.

Abstract

The representation of the human electroencephalogram (EEG) records by neurophysiologists demands standardized time-amplitude scales for their correct conventional interpretation. In a suite of graphical experiments involving scaling affine transformations we have been able to convert electroencephalogram samples corresponding to any particular sleep phase and relaxed wakefulness into each other. We propound a statistical explanation for that finding in terms of data collapse. As a sequel, we determine characteristic time and amplitude scales and outline a possible physical interpretation. An analysis for characteristic times based on lacunarity is also carried out as well as a study of the synchrony between left and right EEG channels.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Brain / physiology*
  • Computer Simulation
  • Data Interpretation, Statistical
  • Diagnosis, Computer-Assisted / methods
  • Electroencephalography / methods*
  • Evoked Potentials / physiology*
  • Humans
  • Models, Neurological*
  • Models, Statistical
  • Sleep Stages / physiology*
  • Wakefulness / physiology*