Dynamics of chaotic maps for modelling the multifractal spectrum of human brain Diffusion Tensor Images

摘要: Abstract The multifractal spectra of 3d Diffusion Tensor Images (DTI) obtained by magnetic resonance imaging the human brain are studied. They shown to deviate substantially from artificial images with same white matter intensity. All spectra, 12 healthy subjects, show common characteristics indicating non-trivial moments To model dynamics chaotic Ikeda map used. DTI for positive q best approximated coupled maps in fully developed regime. coupling constants as small α = 0.01. These results reflect not only tissue architectural complexity brain, but also demonstrate presence and importance between neuron axons. is mirrored deviations negative q-spectra, where rare events dominate. obtain a good agreement q-spectrum dynamics, it enough slightly modify most distributions. representation unique: similar conclusions drawn when other (Tent, Logistic or Henon maps) employed modelling axons network.